Upload folder using huggingface_hub
Browse files- README.md +106 -3
- config.json +95 -0
- configuration_acestep_v15.py +265 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +837 -0
- modeling_acestep_v15_xl_turbo.py +0 -0
- silence_latent.pt +3 -0
README.md
CHANGED
|
@@ -1,3 +1,106 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
pipeline_tag: text-to-audio
|
| 5 |
+
tags:
|
| 6 |
+
- feature-extraction
|
| 7 |
+
- audio
|
| 8 |
+
- music
|
| 9 |
+
- text2music
|
| 10 |
+
- custom_code
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<h1 align="center">ACE-Step 1.5 XL — Turbo (4B DiT)</h1>
|
| 14 |
+
<p align="center">
|
| 15 |
+
<a href="https://ace-step.github.io/ace-step-v1.5.github.io/">Project</a> |
|
| 16 |
+
<a href="https://huggingface.co/collections/ACE-Step/ace-step-15">Hugging Face</a> |
|
| 17 |
+
<a href="https://huggingface.co/spaces/ACE-Step/Ace-Step-v1.5">Space Demo</a> |
|
| 18 |
+
<a href="https://discord.gg/PeWDxrkdj7">Discord</a> |
|
| 19 |
+
<a href="https://arxiv.org/abs/2602.00744">Tech Report</a>
|
| 20 |
+
</p>
|
| 21 |
+
|
| 22 |
+
> ⚠️ **Early Release:** This turbo model is still under evaluation. Quality may improve in future updates.
|
| 23 |
+
|
| 24 |
+
## Model Details
|
| 25 |
+
|
| 26 |
+
This is the **XL (4B) Turbo** variant of ACE-Step 1.5 — a distillation-accelerated model that generates high-quality audio in just 8 steps. Combines the speed of turbo with the quality of the 4B architecture.
|
| 27 |
+
|
| 28 |
+
### XL Architecture
|
| 29 |
+
|
| 30 |
+
| Parameter | Value |
|
| 31 |
+
|-----------|-------|
|
| 32 |
+
| DiT Decoder hidden_size | 2560 |
|
| 33 |
+
| DiT Decoder layers | 32 |
|
| 34 |
+
| DiT Decoder attention heads | 32 |
|
| 35 |
+
| Encoder hidden_size | 2048 |
|
| 36 |
+
| Encoder layers | 8 |
|
| 37 |
+
| Total params | ~4B |
|
| 38 |
+
| Weights size (bf16) | ~18.8 GB |
|
| 39 |
+
| Inference steps | 8 (no CFG, distilled) |
|
| 40 |
+
|
| 41 |
+
### GPU Requirements
|
| 42 |
+
|
| 43 |
+
| VRAM | Support |
|
| 44 |
+
|------|---------|
|
| 45 |
+
| ≥12 GB | With CPU offload + INT8 quantization |
|
| 46 |
+
| ≥16 GB | With CPU offload |
|
| 47 |
+
| ≥20 GB | Without offload (recommended) |
|
| 48 |
+
| ≥24 GB | Full quality (XL + 4B LM) |
|
| 49 |
+
|
| 50 |
+
All LM models (0.6B / 1.7B / 4B) are fully compatible with XL.
|
| 51 |
+
|
| 52 |
+
### Key Features
|
| 53 |
+
|
| 54 |
+
- **💰 Commercial-Ready:** Trained on legally compliant datasets. Generated music can be used for commercial purposes.
|
| 55 |
+
- **📚 Safe Training Data:** Licensed music, royalty-free/public domain, and synthetic (MIDI-to-Audio) data.
|
| 56 |
+
- **⚡ Fast:** 8-step inference — the fastest XL variant.
|
| 57 |
+
- **🔮 Higher Quality:** 4B parameters provide richer audio quality than 2B turbo.
|
| 58 |
+
|
| 59 |
+
## Quick Start
|
| 60 |
+
|
| 61 |
+
```bash
|
| 62 |
+
# Install ACE-Step
|
| 63 |
+
git clone https://github.com/ace-step/ACE-Step-1.5.git
|
| 64 |
+
cd ACE-Step-1.5
|
| 65 |
+
pip install -e .
|
| 66 |
+
|
| 67 |
+
# Download this model
|
| 68 |
+
huggingface-cli download ACE-Step/acestep-v15-xl-turbo --local-dir ./checkpoints/acestep-v15-xl-turbo
|
| 69 |
+
|
| 70 |
+
# Run with Gradio UI
|
| 71 |
+
python acestep --config-path acestep-v15-xl-turbo
|
| 72 |
+
```
|
| 73 |
+
|
| 74 |
+
## Model Zoo
|
| 75 |
+
|
| 76 |
+
### XL (4B) DiT Models
|
| 77 |
+
|
| 78 |
+
| DiT Model | CFG | Steps | Quality | Diversity | Tasks | Hugging Face |
|
| 79 |
+
|-----------|:---:|:-----:|:-------:|:---------:|-------|--------------|
|
| 80 |
+
| `acestep-v15-xl-base` | ✅ | 50 | High | High | All (extract, lego, complete) | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-base) |
|
| 81 |
+
| `acestep-v15-xl-sft` | ✅ | 50 | Very High | Medium | Standard | [Link](https://huggingface.co/ACE-Step/acestep-v15-xl-sft) |
|
| 82 |
+
| **`acestep-v15-xl-turbo`** | ❌ | 8 | Very High | Medium | Standard | This repo |
|
| 83 |
+
|
| 84 |
+
### LM Models (all compatible with XL)
|
| 85 |
+
|
| 86 |
+
| LM Model | Params | Audio Understanding | Composition | Hugging Face |
|
| 87 |
+
|----------|:------:|:-------------------:|:-----------:|--------------|
|
| 88 |
+
| `acestep-5Hz-lm-0.6B` | 0.6B | Medium | Medium | [Link](https://huggingface.co/ACE-Step/acestep-5Hz-lm-0.6B) |
|
| 89 |
+
| `acestep-5Hz-lm-1.7B` | 1.7B | Medium | Medium | Included in main |
|
| 90 |
+
| `acestep-5Hz-lm-4B` | 4B | Strong | Strong | [Link](https://huggingface.co/ACE-Step/acestep-5Hz-lm-4B) |
|
| 91 |
+
|
| 92 |
+
## Acknowledgements
|
| 93 |
+
|
| 94 |
+
This project is co-led by ACE Studio and StepFun.
|
| 95 |
+
|
| 96 |
+
## Citation
|
| 97 |
+
|
| 98 |
+
```BibTeX
|
| 99 |
+
@misc{gong2026acestep,
|
| 100 |
+
title={ACE-Step 1.5: Pushing the Boundaries of Open-Source Music Generation},
|
| 101 |
+
author={Junmin Gong, Yulin Song, Wenxiao Zhao, Sen Wang, Shengyuan Xu, Jing Guo},
|
| 102 |
+
howpublished={\url{https://github.com/ace-step/ACE-Step-1.5}},
|
| 103 |
+
year={2026},
|
| 104 |
+
note={GitHub repository}
|
| 105 |
+
}
|
| 106 |
+
```
|
config.json
ADDED
|
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"AceStepConditionGenerationModel"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "configuration_acestep_v15.AceStepConfig",
|
| 7 |
+
"AutoModel": "modeling_acestep_v15_xl_turbo.AceStepConditionGenerationModel"
|
| 8 |
+
},
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"audio_acoustic_hidden_dim": 64,
|
| 12 |
+
"data_proportion": 0.5,
|
| 13 |
+
"dtype": "bfloat16",
|
| 14 |
+
"encoder_hidden_size": 2048,
|
| 15 |
+
"encoder_intermediate_size": 6144,
|
| 16 |
+
"encoder_num_attention_heads": 16,
|
| 17 |
+
"encoder_num_key_value_heads": 8,
|
| 18 |
+
"fsq_dim": 2048,
|
| 19 |
+
"fsq_input_levels": [
|
| 20 |
+
8,
|
| 21 |
+
8,
|
| 22 |
+
8,
|
| 23 |
+
5,
|
| 24 |
+
5,
|
| 25 |
+
5
|
| 26 |
+
],
|
| 27 |
+
"fsq_input_num_quantizers": 1,
|
| 28 |
+
"head_dim": 128,
|
| 29 |
+
"hidden_act": "silu",
|
| 30 |
+
"hidden_size": 2560,
|
| 31 |
+
"in_channels": 192,
|
| 32 |
+
"initializer_range": 0.02,
|
| 33 |
+
"intermediate_size": 9728,
|
| 34 |
+
"layer_types": [
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"sliding_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"sliding_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"sliding_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"sliding_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"sliding_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"sliding_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"sliding_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"sliding_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"full_attention",
|
| 55 |
+
"sliding_attention",
|
| 56 |
+
"full_attention",
|
| 57 |
+
"sliding_attention",
|
| 58 |
+
"full_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"full_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"full_attention",
|
| 63 |
+
"sliding_attention",
|
| 64 |
+
"full_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"full_attention"
|
| 67 |
+
],
|
| 68 |
+
"max_position_embeddings": 32768,
|
| 69 |
+
"model_type": "acestep",
|
| 70 |
+
"model_version": "turbo",
|
| 71 |
+
"is_turbo": true,
|
| 72 |
+
"num_attention_heads": 32,
|
| 73 |
+
"num_attention_pooler_hidden_layers": 2,
|
| 74 |
+
"num_audio_decoder_hidden_layers": 24,
|
| 75 |
+
"num_hidden_layers": 32,
|
| 76 |
+
"num_key_value_heads": 8,
|
| 77 |
+
"num_lyric_encoder_hidden_layers": 8,
|
| 78 |
+
"num_timbre_encoder_hidden_layers": 4,
|
| 79 |
+
"patch_size": 2,
|
| 80 |
+
"pool_window_size": 5,
|
| 81 |
+
"rms_norm_eps": 1e-06,
|
| 82 |
+
"rope_scaling": null,
|
| 83 |
+
"rope_theta": 1000000,
|
| 84 |
+
"sliding_window": 128,
|
| 85 |
+
"text_hidden_dim": 1024,
|
| 86 |
+
"timbre_fix_frame": 750,
|
| 87 |
+
"timbre_hidden_dim": 64,
|
| 88 |
+
"timestep_mu": -0.4,
|
| 89 |
+
"timestep_sigma": 1.0,
|
| 90 |
+
"transformers_version": "4.57.0",
|
| 91 |
+
"use_cache": true,
|
| 92 |
+
"use_sliding_window": true,
|
| 93 |
+
"vocab_size": 64003,
|
| 94 |
+
"lyric_alignment_layers_config": {"3": [18, 27], "4": [22], "5": [5, 6, 7], "6": [2, 12, 13], "7": [20, 21]}
|
| 95 |
+
}
|
configuration_acestep_v15.py
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# coding=utf-8
|
| 2 |
+
# Copyright 2024 The Qwen team, Alibaba Group and the HuggingFace Inc. team. All rights reserved.
|
| 3 |
+
#
|
| 4 |
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
| 5 |
+
# you may not use this file except in compliance with the License.
|
| 6 |
+
# You may obtain a copy of the License at
|
| 7 |
+
#
|
| 8 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
| 9 |
+
#
|
| 10 |
+
# Unless required by applicable law or agreed to in writing, software
|
| 11 |
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
| 12 |
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 13 |
+
# See the License for the specific language governing permissions and
|
| 14 |
+
# limitations under the License.
|
| 15 |
+
"""AceStep model configuration"""
|
| 16 |
+
|
| 17 |
+
from transformers.configuration_utils import PretrainedConfig, layer_type_validation
|
| 18 |
+
from transformers.modeling_rope_utils import rope_config_validation
|
| 19 |
+
from transformers.utils import logging
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
logger = logging.get_logger(__name__)
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
class AceStepConfig(PretrainedConfig):
|
| 26 |
+
r"""
|
| 27 |
+
This is the configuration class to store the configuration of a [`AceStepModel`]. It is used to instantiate an
|
| 28 |
+
AceStep model according to the specified arguments, defining the model architecture.
|
| 29 |
+
|
| 30 |
+
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
|
| 31 |
+
documentation from [`PretrainedConfig`] for more information.
|
| 32 |
+
|
| 33 |
+
Args:
|
| 34 |
+
vocab_size (`int`, *optional*, defaults to 64003):
|
| 35 |
+
Vocabulary size of the AceStep model. Defines the number of different tokens that can be represented by the
|
| 36 |
+
`inputs_ids` passed when calling the model.
|
| 37 |
+
hidden_size (`int`, *optional*, defaults to 4096):
|
| 38 |
+
Dimension of the hidden representations.
|
| 39 |
+
intermediate_size (`int`, *optional*, defaults to 22016):
|
| 40 |
+
Dimension of the MLP representations.
|
| 41 |
+
num_hidden_layers (`int`, *optional*, defaults to 32):
|
| 42 |
+
Number of hidden layers in the Transformer encoder.
|
| 43 |
+
num_attention_heads (`int`, *optional*, defaults to 32):
|
| 44 |
+
Number of attention heads for each attention layer in the Transformer encoder.
|
| 45 |
+
num_key_value_heads (`int`, *optional*, defaults to 32):
|
| 46 |
+
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
|
| 47 |
+
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
|
| 48 |
+
`num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
|
| 49 |
+
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
|
| 50 |
+
by meanpooling all the original heads within that group. For more details, check out [this
|
| 51 |
+
paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`.
|
| 52 |
+
head_dim (`int`, *optional*, defaults to 128):
|
| 53 |
+
The attention head dimension.
|
| 54 |
+
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
|
| 55 |
+
The non-linear activation function (function or string) in the decoder.
|
| 56 |
+
max_position_embeddings (`int`, *optional*, defaults to 32768):
|
| 57 |
+
The maximum sequence length that this model might ever be used with.
|
| 58 |
+
initializer_range (`float`, *optional*, defaults to 0.02):
|
| 59 |
+
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
|
| 60 |
+
rms_norm_eps (`float`, *optional*, defaults to 1e-06):
|
| 61 |
+
The epsilon used by the rms normalization layers.
|
| 62 |
+
use_cache (`bool`, *optional*, defaults to `True`):
|
| 63 |
+
Whether or not the model should return the last key/values attentions (not used by all models). Only
|
| 64 |
+
relevant if `config.is_decoder=True`.
|
| 65 |
+
tie_word_embeddings (`bool`, *optional*, defaults to `False`):
|
| 66 |
+
Whether the model's input and output word embeddings should be tied.
|
| 67 |
+
rope_theta (`float`, *optional*, defaults to 10000.0):
|
| 68 |
+
The base period of the RoPE embeddings.
|
| 69 |
+
rope_scaling (`Dict`, *optional*):
|
| 70 |
+
Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
|
| 71 |
+
and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
|
| 72 |
+
accordingly.
|
| 73 |
+
Expected contents:
|
| 74 |
+
`rope_type` (`str`):
|
| 75 |
+
The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
|
| 76 |
+
'llama3'], with 'default' being the original RoPE implementation.
|
| 77 |
+
`factor` (`float`, *optional*):
|
| 78 |
+
Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
|
| 79 |
+
most scaling types, a `factor` of x will enable the model to handle sequences of length x *
|
| 80 |
+
original maximum pre-trained length.
|
| 81 |
+
`original_max_position_embeddings` (`int`, *optional*):
|
| 82 |
+
Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
|
| 83 |
+
pretraining.
|
| 84 |
+
`attention_factor` (`float`, *optional*):
|
| 85 |
+
Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
|
| 86 |
+
computation. If unspecified, it defaults to value recommended by the implementation, using the
|
| 87 |
+
`factor` field to infer the suggested value.
|
| 88 |
+
`beta_fast` (`float`, *optional*):
|
| 89 |
+
Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
|
| 90 |
+
ramp function. If unspecified, it defaults to 32.
|
| 91 |
+
`beta_slow` (`float`, *optional*):
|
| 92 |
+
Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
|
| 93 |
+
ramp function. If unspecified, it defaults to 1.
|
| 94 |
+
`short_factor` (`list[float]`, *optional*):
|
| 95 |
+
Only used with 'longrope'. The scaling factor to be applied to short contexts (<
|
| 96 |
+
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
|
| 97 |
+
size divided by the number of attention heads divided by 2
|
| 98 |
+
`long_factor` (`list[float]`, *optional*):
|
| 99 |
+
Only used with 'longrope'. The scaling factor to be applied to long contexts (<
|
| 100 |
+
`original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
|
| 101 |
+
size divided by the number of attention heads divided by 2
|
| 102 |
+
`low_freq_factor` (`float`, *optional*):
|
| 103 |
+
Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
|
| 104 |
+
`high_freq_factor` (`float`, *optional*):
|
| 105 |
+
Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
|
| 106 |
+
attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
|
| 107 |
+
Whether to use a bias in the query, key, value and output projection layers during self-attention.
|
| 108 |
+
use_sliding_window (`bool`, *optional*, defaults to `False`):
|
| 109 |
+
Whether to use sliding window attention.
|
| 110 |
+
sliding_window (`int`, *optional*, defaults to 4096):
|
| 111 |
+
Sliding window attention (SWA) window size. If not specified, will default to `4096`.
|
| 112 |
+
layer_types (`list`, *optional*):
|
| 113 |
+
Attention pattern for each layer.
|
| 114 |
+
attention_dropout (`float`, *optional*, defaults to 0.0):
|
| 115 |
+
The dropout ratio for the attention probabilities.
|
| 116 |
+
|
| 117 |
+
```python
|
| 118 |
+
>>> from acestep.models import AceStepConfig
|
| 119 |
+
|
| 120 |
+
>>> # Initializing an AceStep configuration
|
| 121 |
+
>>> configuration = AceStepConfig()
|
| 122 |
+
|
| 123 |
+
>>> # Initializing a model from the configuration
|
| 124 |
+
>>> model = AceStepConditionGenerationModel(configuration)
|
| 125 |
+
|
| 126 |
+
>>> # Accessing the model configuration
|
| 127 |
+
>>> configuration = model.config
|
| 128 |
+
```"""
|
| 129 |
+
|
| 130 |
+
model_type = "acestep"
|
| 131 |
+
keys_to_ignore_at_inference = ["past_key_values"]
|
| 132 |
+
|
| 133 |
+
# Default tensor parallel plan for the base model
|
| 134 |
+
base_model_tp_plan = {
|
| 135 |
+
"layers.*.self_attn.q_proj": "colwise",
|
| 136 |
+
"layers.*.self_attn.k_proj": "colwise",
|
| 137 |
+
"layers.*.self_attn.v_proj": "colwise",
|
| 138 |
+
"layers.*.self_attn.o_proj": "rowwise",
|
| 139 |
+
"layers.*.mlp.gate_proj": "colwise",
|
| 140 |
+
"layers.*.mlp.up_proj": "colwise",
|
| 141 |
+
"layers.*.mlp.down_proj": "rowwise",
|
| 142 |
+
}
|
| 143 |
+
base_model_pp_plan = {
|
| 144 |
+
"embed_tokens": (["input_ids"], ["inputs_embeds"]),
|
| 145 |
+
"layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
|
| 146 |
+
"norm": (["hidden_states"], ["hidden_states"]),
|
| 147 |
+
}
|
| 148 |
+
def __init__(
|
| 149 |
+
self,
|
| 150 |
+
vocab_size=64003,
|
| 151 |
+
fsq_dim=2048,
|
| 152 |
+
fsq_input_levels=[8, 8, 8, 5, 5, 5],
|
| 153 |
+
fsq_input_num_quantizers=1,
|
| 154 |
+
hidden_size=2048,
|
| 155 |
+
intermediate_size=6144,
|
| 156 |
+
num_hidden_layers=24,
|
| 157 |
+
num_attention_heads=16,
|
| 158 |
+
num_key_value_heads=8,
|
| 159 |
+
head_dim=128,
|
| 160 |
+
hidden_act="silu",
|
| 161 |
+
max_position_embeddings=32768,
|
| 162 |
+
initializer_range=0.02,
|
| 163 |
+
rms_norm_eps=1e-6,
|
| 164 |
+
use_cache=True,
|
| 165 |
+
tie_word_embeddings=True,
|
| 166 |
+
rope_theta=1000000,
|
| 167 |
+
rope_scaling=None,
|
| 168 |
+
attention_bias=False,
|
| 169 |
+
use_sliding_window=True,
|
| 170 |
+
sliding_window=128,
|
| 171 |
+
layer_types=None,
|
| 172 |
+
attention_dropout=0.0,
|
| 173 |
+
num_lyric_encoder_hidden_layers=8,
|
| 174 |
+
audio_acoustic_hidden_dim=64,
|
| 175 |
+
pool_window_size=5,
|
| 176 |
+
text_hidden_dim=1024,
|
| 177 |
+
in_channels=192,
|
| 178 |
+
data_proportion=0.5,
|
| 179 |
+
timestep_mu=-0.4,
|
| 180 |
+
timestep_sigma=1.0,
|
| 181 |
+
timbre_hidden_dim=64,
|
| 182 |
+
num_timbre_encoder_hidden_layers=4,
|
| 183 |
+
timbre_fix_frame=750,
|
| 184 |
+
patch_size=2,
|
| 185 |
+
num_attention_pooler_hidden_layers=2,
|
| 186 |
+
num_audio_decoder_hidden_layers=24,
|
| 187 |
+
lyric_alignment_layers_config=None,
|
| 188 |
+
model_version="turbo",
|
| 189 |
+
**kwargs,
|
| 190 |
+
):
|
| 191 |
+
self.max_position_embeddings = max_position_embeddings
|
| 192 |
+
self.hidden_size = hidden_size
|
| 193 |
+
self.intermediate_size = intermediate_size
|
| 194 |
+
self.num_hidden_layers = num_hidden_layers
|
| 195 |
+
self.num_attention_heads = num_attention_heads
|
| 196 |
+
self.use_sliding_window = use_sliding_window
|
| 197 |
+
self.sliding_window = sliding_window if self.use_sliding_window else None
|
| 198 |
+
|
| 199 |
+
# Text encoder configuration
|
| 200 |
+
self.text_hidden_dim = text_hidden_dim
|
| 201 |
+
|
| 202 |
+
# Lyric encoder configuration
|
| 203 |
+
self.num_lyric_encoder_hidden_layers = num_lyric_encoder_hidden_layers
|
| 204 |
+
self.patch_size = patch_size
|
| 205 |
+
|
| 206 |
+
# Audio semantic token generation configuration
|
| 207 |
+
self.audio_acoustic_hidden_dim = audio_acoustic_hidden_dim
|
| 208 |
+
self.pool_window_size = pool_window_size
|
| 209 |
+
self.in_channels = in_channels
|
| 210 |
+
self.data_proportion = data_proportion
|
| 211 |
+
self.timestep_mu = timestep_mu
|
| 212 |
+
self.timestep_sigma = timestep_sigma
|
| 213 |
+
|
| 214 |
+
# FSQ (Finite Scalar Quantization) configuration
|
| 215 |
+
self.fsq_dim = fsq_dim
|
| 216 |
+
self.fsq_input_levels = fsq_input_levels
|
| 217 |
+
self.fsq_input_num_quantizers = fsq_input_num_quantizers
|
| 218 |
+
|
| 219 |
+
# Timbre encoder configuration
|
| 220 |
+
self.timbre_hidden_dim = timbre_hidden_dim
|
| 221 |
+
self.num_timbre_encoder_hidden_layers = num_timbre_encoder_hidden_layers
|
| 222 |
+
self.timbre_fix_frame = timbre_fix_frame
|
| 223 |
+
self.num_attention_pooler_hidden_layers = num_attention_pooler_hidden_layers
|
| 224 |
+
self.num_audio_decoder_hidden_layers = num_audio_decoder_hidden_layers
|
| 225 |
+
self.lyric_alignment_layers_config = lyric_alignment_layers_config
|
| 226 |
+
self.vocab_size = vocab_size
|
| 227 |
+
|
| 228 |
+
# Backward compatibility: ensure num_key_value_heads is set
|
| 229 |
+
if num_key_value_heads is None:
|
| 230 |
+
num_key_value_heads = num_attention_heads
|
| 231 |
+
|
| 232 |
+
self.num_key_value_heads = num_key_value_heads
|
| 233 |
+
self.head_dim = head_dim
|
| 234 |
+
self.hidden_act = hidden_act
|
| 235 |
+
self.initializer_range = initializer_range
|
| 236 |
+
self.rms_norm_eps = rms_norm_eps
|
| 237 |
+
self.use_cache = use_cache
|
| 238 |
+
self.rope_theta = rope_theta
|
| 239 |
+
self.rope_scaling = rope_scaling
|
| 240 |
+
self.attention_bias = attention_bias
|
| 241 |
+
self.attention_dropout = attention_dropout
|
| 242 |
+
self.model_version = model_version
|
| 243 |
+
|
| 244 |
+
# Validate rotary position embeddings parameters
|
| 245 |
+
# Backward compatibility: if there is a 'type' field, move it to 'rope_type'
|
| 246 |
+
if self.rope_scaling is not None and "type" in self.rope_scaling:
|
| 247 |
+
self.rope_scaling["rope_type"] = self.rope_scaling["type"]
|
| 248 |
+
rope_config_validation(self)
|
| 249 |
+
|
| 250 |
+
self.layer_types = layer_types
|
| 251 |
+
|
| 252 |
+
# Set default layer types if not specified
|
| 253 |
+
if self.layer_types is None:
|
| 254 |
+
self.layer_types = [
|
| 255 |
+
"sliding_attention" if bool((i + 1) % 2) else "full_attention" for i in range(self.num_hidden_layers)
|
| 256 |
+
]
|
| 257 |
+
layer_type_validation(self.layer_types)
|
| 258 |
+
|
| 259 |
+
super().__init__(
|
| 260 |
+
tie_word_embeddings=tie_word_embeddings,
|
| 261 |
+
**kwargs,
|
| 262 |
+
)
|
| 263 |
+
|
| 264 |
+
|
| 265 |
+
__all__ = ["AceStepConfig"]
|
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e54a8413c981797312f53d512dec9944023d03f42f0740cbe86f62582cc27bdc
|
| 3 |
+
size 4986971456
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b45f788b6dc9f15957cfa3d9b403f25f3f64c5211ecaf18dda45b67ea6470664
|
| 3 |
+
size 4986942776
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a4f6bd572eab79cf5f3e4678796f904c6d6c33680a3e6ed09034d6de7cabea4d
|
| 3 |
+
size 4986942808
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a868b05493c73808408f58b33a903371bb2846b53c8b19bd7476991bc8837259
|
| 3 |
+
size 4988483464
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,837 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 4987310726,
|
| 4 |
+
"total_size": 19949242904
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"decoder.condition_embedder.bias": "model-00004-of-00004.safetensors",
|
| 8 |
+
"decoder.condition_embedder.weight": "model-00004-of-00004.safetensors",
|
| 9 |
+
"decoder.layers.0.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"decoder.layers.0.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 11 |
+
"decoder.layers.0.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"decoder.layers.0.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 13 |
+
"decoder.layers.0.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"decoder.layers.0.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"decoder.layers.0.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"decoder.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"decoder.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"decoder.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"decoder.layers.0.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"decoder.layers.0.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 21 |
+
"decoder.layers.0.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"decoder.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"decoder.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"decoder.layers.0.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 25 |
+
"decoder.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"decoder.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 27 |
+
"decoder.layers.0.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 28 |
+
"decoder.layers.1.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 29 |
+
"decoder.layers.1.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 30 |
+
"decoder.layers.1.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 31 |
+
"decoder.layers.1.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 32 |
+
"decoder.layers.1.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 33 |
+
"decoder.layers.1.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 34 |
+
"decoder.layers.1.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 35 |
+
"decoder.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 36 |
+
"decoder.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 37 |
+
"decoder.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 38 |
+
"decoder.layers.1.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 39 |
+
"decoder.layers.1.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 40 |
+
"decoder.layers.1.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 41 |
+
"decoder.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 42 |
+
"decoder.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 43 |
+
"decoder.layers.1.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 44 |
+
"decoder.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 45 |
+
"decoder.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 46 |
+
"decoder.layers.1.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 47 |
+
"decoder.layers.10.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"decoder.layers.10.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"decoder.layers.10.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"decoder.layers.10.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"decoder.layers.10.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"decoder.layers.10.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"decoder.layers.10.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"decoder.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"decoder.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"decoder.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"decoder.layers.10.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"decoder.layers.10.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 59 |
+
"decoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"decoder.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"decoder.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"decoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"decoder.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 64 |
+
"decoder.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"decoder.layers.10.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 66 |
+
"decoder.layers.11.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"decoder.layers.11.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"decoder.layers.11.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"decoder.layers.11.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"decoder.layers.11.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"decoder.layers.11.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"decoder.layers.11.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"decoder.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"decoder.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"decoder.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"decoder.layers.11.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"decoder.layers.11.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 78 |
+
"decoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"decoder.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 80 |
+
"decoder.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"decoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"decoder.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"decoder.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"decoder.layers.11.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"decoder.layers.12.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"decoder.layers.12.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"decoder.layers.12.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 88 |
+
"decoder.layers.12.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 89 |
+
"decoder.layers.12.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 90 |
+
"decoder.layers.12.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 91 |
+
"decoder.layers.12.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 92 |
+
"decoder.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 93 |
+
"decoder.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 94 |
+
"decoder.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 95 |
+
"decoder.layers.12.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 96 |
+
"decoder.layers.12.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 97 |
+
"decoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 98 |
+
"decoder.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 99 |
+
"decoder.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 100 |
+
"decoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 101 |
+
"decoder.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 102 |
+
"decoder.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 103 |
+
"decoder.layers.12.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 104 |
+
"decoder.layers.13.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 105 |
+
"decoder.layers.13.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 106 |
+
"decoder.layers.13.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 107 |
+
"decoder.layers.13.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 108 |
+
"decoder.layers.13.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 109 |
+
"decoder.layers.13.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 110 |
+
"decoder.layers.13.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 111 |
+
"decoder.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 112 |
+
"decoder.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 113 |
+
"decoder.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 114 |
+
"decoder.layers.13.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 115 |
+
"decoder.layers.13.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 116 |
+
"decoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 117 |
+
"decoder.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 118 |
+
"decoder.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 119 |
+
"decoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 120 |
+
"decoder.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 121 |
+
"decoder.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 122 |
+
"decoder.layers.13.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 123 |
+
"decoder.layers.14.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 124 |
+
"decoder.layers.14.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 125 |
+
"decoder.layers.14.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 126 |
+
"decoder.layers.14.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 127 |
+
"decoder.layers.14.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 128 |
+
"decoder.layers.14.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 129 |
+
"decoder.layers.14.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 130 |
+
"decoder.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 131 |
+
"decoder.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 132 |
+
"decoder.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 133 |
+
"decoder.layers.14.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 134 |
+
"decoder.layers.14.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 135 |
+
"decoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 136 |
+
"decoder.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 137 |
+
"decoder.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 138 |
+
"decoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 139 |
+
"decoder.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 140 |
+
"decoder.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 141 |
+
"decoder.layers.14.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 142 |
+
"decoder.layers.15.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 143 |
+
"decoder.layers.15.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 144 |
+
"decoder.layers.15.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 145 |
+
"decoder.layers.15.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 146 |
+
"decoder.layers.15.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 147 |
+
"decoder.layers.15.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 148 |
+
"decoder.layers.15.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 149 |
+
"decoder.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 150 |
+
"decoder.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 151 |
+
"decoder.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 152 |
+
"decoder.layers.15.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 153 |
+
"decoder.layers.15.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 154 |
+
"decoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 155 |
+
"decoder.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 156 |
+
"decoder.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 157 |
+
"decoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 158 |
+
"decoder.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 159 |
+
"decoder.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 160 |
+
"decoder.layers.15.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 161 |
+
"decoder.layers.16.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 162 |
+
"decoder.layers.16.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 163 |
+
"decoder.layers.16.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 164 |
+
"decoder.layers.16.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 165 |
+
"decoder.layers.16.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 166 |
+
"decoder.layers.16.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 167 |
+
"decoder.layers.16.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 168 |
+
"decoder.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 169 |
+
"decoder.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 170 |
+
"decoder.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 171 |
+
"decoder.layers.16.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 172 |
+
"decoder.layers.16.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 173 |
+
"decoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 174 |
+
"decoder.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 175 |
+
"decoder.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 176 |
+
"decoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 177 |
+
"decoder.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 178 |
+
"decoder.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 179 |
+
"decoder.layers.16.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 180 |
+
"decoder.layers.17.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 181 |
+
"decoder.layers.17.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 182 |
+
"decoder.layers.17.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 183 |
+
"decoder.layers.17.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 184 |
+
"decoder.layers.17.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 185 |
+
"decoder.layers.17.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 186 |
+
"decoder.layers.17.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 187 |
+
"decoder.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 188 |
+
"decoder.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 189 |
+
"decoder.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 190 |
+
"decoder.layers.17.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 191 |
+
"decoder.layers.17.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 192 |
+
"decoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 193 |
+
"decoder.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 194 |
+
"decoder.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 195 |
+
"decoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 196 |
+
"decoder.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 197 |
+
"decoder.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 198 |
+
"decoder.layers.17.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 199 |
+
"decoder.layers.18.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 200 |
+
"decoder.layers.18.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 201 |
+
"decoder.layers.18.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 202 |
+
"decoder.layers.18.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 203 |
+
"decoder.layers.18.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 204 |
+
"decoder.layers.18.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 205 |
+
"decoder.layers.18.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 206 |
+
"decoder.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 207 |
+
"decoder.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 208 |
+
"decoder.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 209 |
+
"decoder.layers.18.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 210 |
+
"decoder.layers.18.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 211 |
+
"decoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 212 |
+
"decoder.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 213 |
+
"decoder.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 214 |
+
"decoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 215 |
+
"decoder.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 216 |
+
"decoder.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 217 |
+
"decoder.layers.18.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 218 |
+
"decoder.layers.19.cross_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 219 |
+
"decoder.layers.19.cross_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 220 |
+
"decoder.layers.19.cross_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 221 |
+
"decoder.layers.19.cross_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 222 |
+
"decoder.layers.19.cross_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 223 |
+
"decoder.layers.19.cross_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 224 |
+
"decoder.layers.19.cross_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 225 |
+
"decoder.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 226 |
+
"decoder.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 227 |
+
"decoder.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 228 |
+
"decoder.layers.19.mlp_norm.weight": "model-00002-of-00004.safetensors",
|
| 229 |
+
"decoder.layers.19.scale_shift_table": "model-00002-of-00004.safetensors",
|
| 230 |
+
"decoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
|
| 231 |
+
"decoder.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 232 |
+
"decoder.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 233 |
+
"decoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
|
| 234 |
+
"decoder.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 235 |
+
"decoder.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 236 |
+
"decoder.layers.19.self_attn_norm.weight": "model-00002-of-00004.safetensors",
|
| 237 |
+
"decoder.layers.2.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 238 |
+
"decoder.layers.2.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 239 |
+
"decoder.layers.2.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 240 |
+
"decoder.layers.2.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 241 |
+
"decoder.layers.2.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 242 |
+
"decoder.layers.2.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 243 |
+
"decoder.layers.2.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 244 |
+
"decoder.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 245 |
+
"decoder.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 246 |
+
"decoder.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 247 |
+
"decoder.layers.2.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 248 |
+
"decoder.layers.2.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 249 |
+
"decoder.layers.2.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 250 |
+
"decoder.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 251 |
+
"decoder.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 252 |
+
"decoder.layers.2.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 253 |
+
"decoder.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 254 |
+
"decoder.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 255 |
+
"decoder.layers.2.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 256 |
+
"decoder.layers.20.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 257 |
+
"decoder.layers.20.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 258 |
+
"decoder.layers.20.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 259 |
+
"decoder.layers.20.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 260 |
+
"decoder.layers.20.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 261 |
+
"decoder.layers.20.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 262 |
+
"decoder.layers.20.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 263 |
+
"decoder.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 264 |
+
"decoder.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 265 |
+
"decoder.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 266 |
+
"decoder.layers.20.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 267 |
+
"decoder.layers.20.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 268 |
+
"decoder.layers.20.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 269 |
+
"decoder.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 270 |
+
"decoder.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 271 |
+
"decoder.layers.20.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 272 |
+
"decoder.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 273 |
+
"decoder.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 274 |
+
"decoder.layers.20.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 275 |
+
"decoder.layers.21.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 276 |
+
"decoder.layers.21.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 277 |
+
"decoder.layers.21.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 278 |
+
"decoder.layers.21.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 279 |
+
"decoder.layers.21.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 280 |
+
"decoder.layers.21.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 281 |
+
"decoder.layers.21.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 282 |
+
"decoder.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 283 |
+
"decoder.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 284 |
+
"decoder.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 285 |
+
"decoder.layers.21.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 286 |
+
"decoder.layers.21.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 287 |
+
"decoder.layers.21.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 288 |
+
"decoder.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 289 |
+
"decoder.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 290 |
+
"decoder.layers.21.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 291 |
+
"decoder.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 292 |
+
"decoder.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 293 |
+
"decoder.layers.21.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 294 |
+
"decoder.layers.22.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 295 |
+
"decoder.layers.22.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 296 |
+
"decoder.layers.22.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 297 |
+
"decoder.layers.22.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 298 |
+
"decoder.layers.22.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 299 |
+
"decoder.layers.22.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 300 |
+
"decoder.layers.22.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 301 |
+
"decoder.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 302 |
+
"decoder.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 303 |
+
"decoder.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 304 |
+
"decoder.layers.22.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 305 |
+
"decoder.layers.22.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 306 |
+
"decoder.layers.22.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 307 |
+
"decoder.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 308 |
+
"decoder.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 309 |
+
"decoder.layers.22.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 310 |
+
"decoder.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 311 |
+
"decoder.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 312 |
+
"decoder.layers.22.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 313 |
+
"decoder.layers.23.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 314 |
+
"decoder.layers.23.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 315 |
+
"decoder.layers.23.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 316 |
+
"decoder.layers.23.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 317 |
+
"decoder.layers.23.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 318 |
+
"decoder.layers.23.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 319 |
+
"decoder.layers.23.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 320 |
+
"decoder.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 321 |
+
"decoder.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 322 |
+
"decoder.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 323 |
+
"decoder.layers.23.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 324 |
+
"decoder.layers.23.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 325 |
+
"decoder.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 326 |
+
"decoder.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 327 |
+
"decoder.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 328 |
+
"decoder.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 329 |
+
"decoder.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 330 |
+
"decoder.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 331 |
+
"decoder.layers.23.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 332 |
+
"decoder.layers.24.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 333 |
+
"decoder.layers.24.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 334 |
+
"decoder.layers.24.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 335 |
+
"decoder.layers.24.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 336 |
+
"decoder.layers.24.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 337 |
+
"decoder.layers.24.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 338 |
+
"decoder.layers.24.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 339 |
+
"decoder.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 340 |
+
"decoder.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 341 |
+
"decoder.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 342 |
+
"decoder.layers.24.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 343 |
+
"decoder.layers.24.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 344 |
+
"decoder.layers.24.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 345 |
+
"decoder.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 346 |
+
"decoder.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 347 |
+
"decoder.layers.24.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 348 |
+
"decoder.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 349 |
+
"decoder.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 350 |
+
"decoder.layers.24.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 351 |
+
"decoder.layers.25.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 352 |
+
"decoder.layers.25.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 353 |
+
"decoder.layers.25.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 354 |
+
"decoder.layers.25.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 355 |
+
"decoder.layers.25.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 356 |
+
"decoder.layers.25.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 357 |
+
"decoder.layers.25.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 358 |
+
"decoder.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 359 |
+
"decoder.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 360 |
+
"decoder.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 361 |
+
"decoder.layers.25.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 362 |
+
"decoder.layers.25.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 363 |
+
"decoder.layers.25.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 364 |
+
"decoder.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 365 |
+
"decoder.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 366 |
+
"decoder.layers.25.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 367 |
+
"decoder.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 368 |
+
"decoder.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 369 |
+
"decoder.layers.25.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 370 |
+
"decoder.layers.26.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 371 |
+
"decoder.layers.26.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 372 |
+
"decoder.layers.26.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 373 |
+
"decoder.layers.26.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 374 |
+
"decoder.layers.26.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 375 |
+
"decoder.layers.26.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 376 |
+
"decoder.layers.26.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 377 |
+
"decoder.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 378 |
+
"decoder.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 379 |
+
"decoder.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 380 |
+
"decoder.layers.26.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 381 |
+
"decoder.layers.26.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 382 |
+
"decoder.layers.26.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 383 |
+
"decoder.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 384 |
+
"decoder.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 385 |
+
"decoder.layers.26.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 386 |
+
"decoder.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 387 |
+
"decoder.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 388 |
+
"decoder.layers.26.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 389 |
+
"decoder.layers.27.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 390 |
+
"decoder.layers.27.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 391 |
+
"decoder.layers.27.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 392 |
+
"decoder.layers.27.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 393 |
+
"decoder.layers.27.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 394 |
+
"decoder.layers.27.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 395 |
+
"decoder.layers.27.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 396 |
+
"decoder.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 397 |
+
"decoder.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 398 |
+
"decoder.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 399 |
+
"decoder.layers.27.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 400 |
+
"decoder.layers.27.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 401 |
+
"decoder.layers.27.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 402 |
+
"decoder.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 403 |
+
"decoder.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 404 |
+
"decoder.layers.27.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 405 |
+
"decoder.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 406 |
+
"decoder.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 407 |
+
"decoder.layers.27.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 408 |
+
"decoder.layers.28.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 409 |
+
"decoder.layers.28.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 410 |
+
"decoder.layers.28.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 411 |
+
"decoder.layers.28.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 412 |
+
"decoder.layers.28.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 413 |
+
"decoder.layers.28.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 414 |
+
"decoder.layers.28.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 415 |
+
"decoder.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 416 |
+
"decoder.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 417 |
+
"decoder.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 418 |
+
"decoder.layers.28.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 419 |
+
"decoder.layers.28.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 420 |
+
"decoder.layers.28.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 421 |
+
"decoder.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 422 |
+
"decoder.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 423 |
+
"decoder.layers.28.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 424 |
+
"decoder.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 425 |
+
"decoder.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 426 |
+
"decoder.layers.28.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 427 |
+
"decoder.layers.29.cross_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 428 |
+
"decoder.layers.29.cross_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 429 |
+
"decoder.layers.29.cross_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 430 |
+
"decoder.layers.29.cross_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 431 |
+
"decoder.layers.29.cross_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 432 |
+
"decoder.layers.29.cross_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 433 |
+
"decoder.layers.29.cross_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 434 |
+
"decoder.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 435 |
+
"decoder.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 436 |
+
"decoder.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 437 |
+
"decoder.layers.29.mlp_norm.weight": "model-00003-of-00004.safetensors",
|
| 438 |
+
"decoder.layers.29.scale_shift_table": "model-00003-of-00004.safetensors",
|
| 439 |
+
"decoder.layers.29.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
|
| 440 |
+
"decoder.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 441 |
+
"decoder.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 442 |
+
"decoder.layers.29.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
|
| 443 |
+
"decoder.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 444 |
+
"decoder.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 445 |
+
"decoder.layers.29.self_attn_norm.weight": "model-00003-of-00004.safetensors",
|
| 446 |
+
"decoder.layers.3.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 447 |
+
"decoder.layers.3.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 448 |
+
"decoder.layers.3.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 449 |
+
"decoder.layers.3.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 450 |
+
"decoder.layers.3.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 451 |
+
"decoder.layers.3.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 452 |
+
"decoder.layers.3.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 453 |
+
"decoder.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 454 |
+
"decoder.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 455 |
+
"decoder.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 456 |
+
"decoder.layers.3.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 457 |
+
"decoder.layers.3.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 458 |
+
"decoder.layers.3.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 459 |
+
"decoder.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 460 |
+
"decoder.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 461 |
+
"decoder.layers.3.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 462 |
+
"decoder.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 463 |
+
"decoder.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 464 |
+
"decoder.layers.3.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 465 |
+
"decoder.layers.30.cross_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 466 |
+
"decoder.layers.30.cross_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 467 |
+
"decoder.layers.30.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 468 |
+
"decoder.layers.30.cross_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 469 |
+
"decoder.layers.30.cross_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 470 |
+
"decoder.layers.30.cross_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 471 |
+
"decoder.layers.30.cross_attn_norm.weight": "model-00004-of-00004.safetensors",
|
| 472 |
+
"decoder.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 473 |
+
"decoder.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 474 |
+
"decoder.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 475 |
+
"decoder.layers.30.mlp_norm.weight": "model-00004-of-00004.safetensors",
|
| 476 |
+
"decoder.layers.30.scale_shift_table": "model-00004-of-00004.safetensors",
|
| 477 |
+
"decoder.layers.30.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 478 |
+
"decoder.layers.30.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 479 |
+
"decoder.layers.30.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 480 |
+
"decoder.layers.30.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 481 |
+
"decoder.layers.30.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 482 |
+
"decoder.layers.30.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 483 |
+
"decoder.layers.30.self_attn_norm.weight": "model-00004-of-00004.safetensors",
|
| 484 |
+
"decoder.layers.31.cross_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 485 |
+
"decoder.layers.31.cross_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 486 |
+
"decoder.layers.31.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 487 |
+
"decoder.layers.31.cross_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 488 |
+
"decoder.layers.31.cross_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 489 |
+
"decoder.layers.31.cross_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 490 |
+
"decoder.layers.31.cross_attn_norm.weight": "model-00004-of-00004.safetensors",
|
| 491 |
+
"decoder.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 492 |
+
"decoder.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 493 |
+
"decoder.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 494 |
+
"decoder.layers.31.mlp_norm.weight": "model-00004-of-00004.safetensors",
|
| 495 |
+
"decoder.layers.31.scale_shift_table": "model-00004-of-00004.safetensors",
|
| 496 |
+
"decoder.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 497 |
+
"decoder.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 498 |
+
"decoder.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 499 |
+
"decoder.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 500 |
+
"decoder.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 501 |
+
"decoder.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 502 |
+
"decoder.layers.31.self_attn_norm.weight": "model-00004-of-00004.safetensors",
|
| 503 |
+
"decoder.layers.4.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 504 |
+
"decoder.layers.4.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 505 |
+
"decoder.layers.4.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 506 |
+
"decoder.layers.4.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 507 |
+
"decoder.layers.4.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 508 |
+
"decoder.layers.4.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 509 |
+
"decoder.layers.4.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 510 |
+
"decoder.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 511 |
+
"decoder.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 512 |
+
"decoder.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 513 |
+
"decoder.layers.4.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 514 |
+
"decoder.layers.4.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 515 |
+
"decoder.layers.4.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 516 |
+
"decoder.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 517 |
+
"decoder.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 518 |
+
"decoder.layers.4.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 519 |
+
"decoder.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 520 |
+
"decoder.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 521 |
+
"decoder.layers.4.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 522 |
+
"decoder.layers.5.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 523 |
+
"decoder.layers.5.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 524 |
+
"decoder.layers.5.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 525 |
+
"decoder.layers.5.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 526 |
+
"decoder.layers.5.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 527 |
+
"decoder.layers.5.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 528 |
+
"decoder.layers.5.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 529 |
+
"decoder.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 530 |
+
"decoder.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 531 |
+
"decoder.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 532 |
+
"decoder.layers.5.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 533 |
+
"decoder.layers.5.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 534 |
+
"decoder.layers.5.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 535 |
+
"decoder.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 536 |
+
"decoder.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 537 |
+
"decoder.layers.5.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 538 |
+
"decoder.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 539 |
+
"decoder.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 540 |
+
"decoder.layers.5.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 541 |
+
"decoder.layers.6.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 542 |
+
"decoder.layers.6.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 543 |
+
"decoder.layers.6.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 544 |
+
"decoder.layers.6.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 545 |
+
"decoder.layers.6.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 546 |
+
"decoder.layers.6.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 547 |
+
"decoder.layers.6.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 548 |
+
"decoder.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 549 |
+
"decoder.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 550 |
+
"decoder.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 551 |
+
"decoder.layers.6.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 552 |
+
"decoder.layers.6.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 553 |
+
"decoder.layers.6.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 554 |
+
"decoder.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 555 |
+
"decoder.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 556 |
+
"decoder.layers.6.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 557 |
+
"decoder.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 558 |
+
"decoder.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 559 |
+
"decoder.layers.6.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 560 |
+
"decoder.layers.7.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 561 |
+
"decoder.layers.7.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 562 |
+
"decoder.layers.7.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 563 |
+
"decoder.layers.7.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 564 |
+
"decoder.layers.7.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 565 |
+
"decoder.layers.7.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 566 |
+
"decoder.layers.7.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 567 |
+
"decoder.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 568 |
+
"decoder.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 569 |
+
"decoder.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 570 |
+
"decoder.layers.7.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 571 |
+
"decoder.layers.7.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 572 |
+
"decoder.layers.7.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 573 |
+
"decoder.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 574 |
+
"decoder.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 575 |
+
"decoder.layers.7.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 576 |
+
"decoder.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 577 |
+
"decoder.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 578 |
+
"decoder.layers.7.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 579 |
+
"decoder.layers.8.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 580 |
+
"decoder.layers.8.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 581 |
+
"decoder.layers.8.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 582 |
+
"decoder.layers.8.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 583 |
+
"decoder.layers.8.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 584 |
+
"decoder.layers.8.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 585 |
+
"decoder.layers.8.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 586 |
+
"decoder.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 587 |
+
"decoder.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 588 |
+
"decoder.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 589 |
+
"decoder.layers.8.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 590 |
+
"decoder.layers.8.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 591 |
+
"decoder.layers.8.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 592 |
+
"decoder.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 593 |
+
"decoder.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 594 |
+
"decoder.layers.8.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 595 |
+
"decoder.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 596 |
+
"decoder.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 597 |
+
"decoder.layers.8.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 598 |
+
"decoder.layers.9.cross_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 599 |
+
"decoder.layers.9.cross_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 600 |
+
"decoder.layers.9.cross_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 601 |
+
"decoder.layers.9.cross_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 602 |
+
"decoder.layers.9.cross_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 603 |
+
"decoder.layers.9.cross_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 604 |
+
"decoder.layers.9.cross_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 605 |
+
"decoder.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 606 |
+
"decoder.layers.9.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 607 |
+
"decoder.layers.9.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 608 |
+
"decoder.layers.9.mlp_norm.weight": "model-00001-of-00004.safetensors",
|
| 609 |
+
"decoder.layers.9.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 610 |
+
"decoder.layers.9.self_attn.k_norm.weight": "model-00001-of-00004.safetensors",
|
| 611 |
+
"decoder.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 612 |
+
"decoder.layers.9.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 613 |
+
"decoder.layers.9.self_attn.q_norm.weight": "model-00001-of-00004.safetensors",
|
| 614 |
+
"decoder.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 615 |
+
"decoder.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 616 |
+
"decoder.layers.9.self_attn_norm.weight": "model-00001-of-00004.safetensors",
|
| 617 |
+
"decoder.norm_out.weight": "model-00004-of-00004.safetensors",
|
| 618 |
+
"decoder.proj_in.1.bias": "model-00004-of-00004.safetensors",
|
| 619 |
+
"decoder.proj_in.1.weight": "model-00004-of-00004.safetensors",
|
| 620 |
+
"decoder.proj_out.1.bias": "model-00004-of-00004.safetensors",
|
| 621 |
+
"decoder.proj_out.1.weight": "model-00004-of-00004.safetensors",
|
| 622 |
+
"decoder.scale_shift_table": "model-00001-of-00004.safetensors",
|
| 623 |
+
"decoder.time_embed.linear_1.bias": "model-00004-of-00004.safetensors",
|
| 624 |
+
"decoder.time_embed.linear_1.weight": "model-00004-of-00004.safetensors",
|
| 625 |
+
"decoder.time_embed.linear_2.bias": "model-00004-of-00004.safetensors",
|
| 626 |
+
"decoder.time_embed.linear_2.weight": "model-00004-of-00004.safetensors",
|
| 627 |
+
"decoder.time_embed.time_proj.bias": "model-00004-of-00004.safetensors",
|
| 628 |
+
"decoder.time_embed.time_proj.weight": "model-00004-of-00004.safetensors",
|
| 629 |
+
"decoder.time_embed_r.linear_1.bias": "model-00004-of-00004.safetensors",
|
| 630 |
+
"decoder.time_embed_r.linear_1.weight": "model-00004-of-00004.safetensors",
|
| 631 |
+
"decoder.time_embed_r.linear_2.bias": "model-00004-of-00004.safetensors",
|
| 632 |
+
"decoder.time_embed_r.linear_2.weight": "model-00004-of-00004.safetensors",
|
| 633 |
+
"decoder.time_embed_r.time_proj.bias": "model-00004-of-00004.safetensors",
|
| 634 |
+
"decoder.time_embed_r.time_proj.weight": "model-00004-of-00004.safetensors",
|
| 635 |
+
"detokenizer.embed_tokens.bias": "model-00004-of-00004.safetensors",
|
| 636 |
+
"detokenizer.embed_tokens.weight": "model-00004-of-00004.safetensors",
|
| 637 |
+
"detokenizer.layers.0.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 638 |
+
"detokenizer.layers.0.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 639 |
+
"detokenizer.layers.0.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 640 |
+
"detokenizer.layers.0.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 641 |
+
"detokenizer.layers.0.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 642 |
+
"detokenizer.layers.0.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 643 |
+
"detokenizer.layers.0.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 644 |
+
"detokenizer.layers.0.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 645 |
+
"detokenizer.layers.0.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 646 |
+
"detokenizer.layers.0.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 647 |
+
"detokenizer.layers.0.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 648 |
+
"detokenizer.layers.1.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 649 |
+
"detokenizer.layers.1.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 650 |
+
"detokenizer.layers.1.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 651 |
+
"detokenizer.layers.1.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 652 |
+
"detokenizer.layers.1.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 653 |
+
"detokenizer.layers.1.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 654 |
+
"detokenizer.layers.1.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 655 |
+
"detokenizer.layers.1.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 656 |
+
"detokenizer.layers.1.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 657 |
+
"detokenizer.layers.1.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 658 |
+
"detokenizer.layers.1.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 659 |
+
"detokenizer.norm.weight": "model-00004-of-00004.safetensors",
|
| 660 |
+
"detokenizer.proj_out.bias": "model-00004-of-00004.safetensors",
|
| 661 |
+
"detokenizer.proj_out.weight": "model-00004-of-00004.safetensors",
|
| 662 |
+
"detokenizer.special_tokens": "model-00004-of-00004.safetensors",
|
| 663 |
+
"encoder.lyric_encoder.embed_tokens.bias": "model-00004-of-00004.safetensors",
|
| 664 |
+
"encoder.lyric_encoder.embed_tokens.weight": "model-00004-of-00004.safetensors",
|
| 665 |
+
"encoder.lyric_encoder.layers.0.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 666 |
+
"encoder.lyric_encoder.layers.0.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 667 |
+
"encoder.lyric_encoder.layers.0.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 668 |
+
"encoder.lyric_encoder.layers.0.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 669 |
+
"encoder.lyric_encoder.layers.0.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 670 |
+
"encoder.lyric_encoder.layers.0.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 671 |
+
"encoder.lyric_encoder.layers.0.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 672 |
+
"encoder.lyric_encoder.layers.0.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 673 |
+
"encoder.lyric_encoder.layers.0.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 674 |
+
"encoder.lyric_encoder.layers.0.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 675 |
+
"encoder.lyric_encoder.layers.0.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 676 |
+
"encoder.lyric_encoder.layers.1.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 677 |
+
"encoder.lyric_encoder.layers.1.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 678 |
+
"encoder.lyric_encoder.layers.1.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 679 |
+
"encoder.lyric_encoder.layers.1.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 680 |
+
"encoder.lyric_encoder.layers.1.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 681 |
+
"encoder.lyric_encoder.layers.1.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 682 |
+
"encoder.lyric_encoder.layers.1.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 683 |
+
"encoder.lyric_encoder.layers.1.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 684 |
+
"encoder.lyric_encoder.layers.1.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 685 |
+
"encoder.lyric_encoder.layers.1.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 686 |
+
"encoder.lyric_encoder.layers.1.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 687 |
+
"encoder.lyric_encoder.layers.2.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 688 |
+
"encoder.lyric_encoder.layers.2.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 689 |
+
"encoder.lyric_encoder.layers.2.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 690 |
+
"encoder.lyric_encoder.layers.2.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 691 |
+
"encoder.lyric_encoder.layers.2.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 692 |
+
"encoder.lyric_encoder.layers.2.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 693 |
+
"encoder.lyric_encoder.layers.2.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 694 |
+
"encoder.lyric_encoder.layers.2.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 695 |
+
"encoder.lyric_encoder.layers.2.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 696 |
+
"encoder.lyric_encoder.layers.2.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 697 |
+
"encoder.lyric_encoder.layers.2.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 698 |
+
"encoder.lyric_encoder.layers.3.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 699 |
+
"encoder.lyric_encoder.layers.3.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 700 |
+
"encoder.lyric_encoder.layers.3.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 701 |
+
"encoder.lyric_encoder.layers.3.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 702 |
+
"encoder.lyric_encoder.layers.3.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 703 |
+
"encoder.lyric_encoder.layers.3.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 704 |
+
"encoder.lyric_encoder.layers.3.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 705 |
+
"encoder.lyric_encoder.layers.3.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 706 |
+
"encoder.lyric_encoder.layers.3.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 707 |
+
"encoder.lyric_encoder.layers.3.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 708 |
+
"encoder.lyric_encoder.layers.3.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 709 |
+
"encoder.lyric_encoder.layers.4.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 710 |
+
"encoder.lyric_encoder.layers.4.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 711 |
+
"encoder.lyric_encoder.layers.4.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 712 |
+
"encoder.lyric_encoder.layers.4.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 713 |
+
"encoder.lyric_encoder.layers.4.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 714 |
+
"encoder.lyric_encoder.layers.4.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 715 |
+
"encoder.lyric_encoder.layers.4.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 716 |
+
"encoder.lyric_encoder.layers.4.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 717 |
+
"encoder.lyric_encoder.layers.4.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 718 |
+
"encoder.lyric_encoder.layers.4.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 719 |
+
"encoder.lyric_encoder.layers.4.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 720 |
+
"encoder.lyric_encoder.layers.5.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 721 |
+
"encoder.lyric_encoder.layers.5.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 722 |
+
"encoder.lyric_encoder.layers.5.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 723 |
+
"encoder.lyric_encoder.layers.5.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 724 |
+
"encoder.lyric_encoder.layers.5.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 725 |
+
"encoder.lyric_encoder.layers.5.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 726 |
+
"encoder.lyric_encoder.layers.5.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 727 |
+
"encoder.lyric_encoder.layers.5.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 728 |
+
"encoder.lyric_encoder.layers.5.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 729 |
+
"encoder.lyric_encoder.layers.5.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 730 |
+
"encoder.lyric_encoder.layers.5.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 731 |
+
"encoder.lyric_encoder.layers.6.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 732 |
+
"encoder.lyric_encoder.layers.6.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 733 |
+
"encoder.lyric_encoder.layers.6.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 734 |
+
"encoder.lyric_encoder.layers.6.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 735 |
+
"encoder.lyric_encoder.layers.6.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 736 |
+
"encoder.lyric_encoder.layers.6.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 737 |
+
"encoder.lyric_encoder.layers.6.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 738 |
+
"encoder.lyric_encoder.layers.6.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 739 |
+
"encoder.lyric_encoder.layers.6.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 740 |
+
"encoder.lyric_encoder.layers.6.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 741 |
+
"encoder.lyric_encoder.layers.6.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 742 |
+
"encoder.lyric_encoder.layers.7.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 743 |
+
"encoder.lyric_encoder.layers.7.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 744 |
+
"encoder.lyric_encoder.layers.7.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 745 |
+
"encoder.lyric_encoder.layers.7.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 746 |
+
"encoder.lyric_encoder.layers.7.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 747 |
+
"encoder.lyric_encoder.layers.7.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 748 |
+
"encoder.lyric_encoder.layers.7.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 749 |
+
"encoder.lyric_encoder.layers.7.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 750 |
+
"encoder.lyric_encoder.layers.7.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 751 |
+
"encoder.lyric_encoder.layers.7.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 752 |
+
"encoder.lyric_encoder.layers.7.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 753 |
+
"encoder.lyric_encoder.norm.weight": "model-00004-of-00004.safetensors",
|
| 754 |
+
"encoder.text_projector.weight": "model-00004-of-00004.safetensors",
|
| 755 |
+
"encoder.timbre_encoder.embed_tokens.bias": "model-00004-of-00004.safetensors",
|
| 756 |
+
"encoder.timbre_encoder.embed_tokens.weight": "model-00004-of-00004.safetensors",
|
| 757 |
+
"encoder.timbre_encoder.layers.0.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 758 |
+
"encoder.timbre_encoder.layers.0.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 759 |
+
"encoder.timbre_encoder.layers.0.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 760 |
+
"encoder.timbre_encoder.layers.0.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 761 |
+
"encoder.timbre_encoder.layers.0.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 762 |
+
"encoder.timbre_encoder.layers.0.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 763 |
+
"encoder.timbre_encoder.layers.0.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 764 |
+
"encoder.timbre_encoder.layers.0.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 765 |
+
"encoder.timbre_encoder.layers.0.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 766 |
+
"encoder.timbre_encoder.layers.0.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 767 |
+
"encoder.timbre_encoder.layers.0.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 768 |
+
"encoder.timbre_encoder.layers.1.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 769 |
+
"encoder.timbre_encoder.layers.1.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 770 |
+
"encoder.timbre_encoder.layers.1.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 771 |
+
"encoder.timbre_encoder.layers.1.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 772 |
+
"encoder.timbre_encoder.layers.1.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 773 |
+
"encoder.timbre_encoder.layers.1.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 774 |
+
"encoder.timbre_encoder.layers.1.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 775 |
+
"encoder.timbre_encoder.layers.1.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 776 |
+
"encoder.timbre_encoder.layers.1.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 777 |
+
"encoder.timbre_encoder.layers.1.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 778 |
+
"encoder.timbre_encoder.layers.1.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 779 |
+
"encoder.timbre_encoder.layers.2.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 780 |
+
"encoder.timbre_encoder.layers.2.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 781 |
+
"encoder.timbre_encoder.layers.2.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 782 |
+
"encoder.timbre_encoder.layers.2.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 783 |
+
"encoder.timbre_encoder.layers.2.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 784 |
+
"encoder.timbre_encoder.layers.2.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 785 |
+
"encoder.timbre_encoder.layers.2.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 786 |
+
"encoder.timbre_encoder.layers.2.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 787 |
+
"encoder.timbre_encoder.layers.2.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 788 |
+
"encoder.timbre_encoder.layers.2.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 789 |
+
"encoder.timbre_encoder.layers.2.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 790 |
+
"encoder.timbre_encoder.layers.3.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 791 |
+
"encoder.timbre_encoder.layers.3.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 792 |
+
"encoder.timbre_encoder.layers.3.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 793 |
+
"encoder.timbre_encoder.layers.3.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 794 |
+
"encoder.timbre_encoder.layers.3.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 795 |
+
"encoder.timbre_encoder.layers.3.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 796 |
+
"encoder.timbre_encoder.layers.3.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 797 |
+
"encoder.timbre_encoder.layers.3.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 798 |
+
"encoder.timbre_encoder.layers.3.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 799 |
+
"encoder.timbre_encoder.layers.3.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 800 |
+
"encoder.timbre_encoder.layers.3.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 801 |
+
"encoder.timbre_encoder.norm.weight": "model-00004-of-00004.safetensors",
|
| 802 |
+
"encoder.timbre_encoder.special_token": "model-00004-of-00004.safetensors",
|
| 803 |
+
"null_condition_emb": "model-00001-of-00004.safetensors",
|
| 804 |
+
"tokenizer.attention_pooler.embed_tokens.bias": "model-00004-of-00004.safetensors",
|
| 805 |
+
"tokenizer.attention_pooler.embed_tokens.weight": "model-00004-of-00004.safetensors",
|
| 806 |
+
"tokenizer.attention_pooler.layers.0.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 807 |
+
"tokenizer.attention_pooler.layers.0.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 808 |
+
"tokenizer.attention_pooler.layers.0.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 809 |
+
"tokenizer.attention_pooler.layers.0.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 810 |
+
"tokenizer.attention_pooler.layers.0.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 811 |
+
"tokenizer.attention_pooler.layers.0.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 812 |
+
"tokenizer.attention_pooler.layers.0.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 813 |
+
"tokenizer.attention_pooler.layers.0.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 814 |
+
"tokenizer.attention_pooler.layers.0.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 815 |
+
"tokenizer.attention_pooler.layers.0.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 816 |
+
"tokenizer.attention_pooler.layers.0.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 817 |
+
"tokenizer.attention_pooler.layers.1.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 818 |
+
"tokenizer.attention_pooler.layers.1.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 819 |
+
"tokenizer.attention_pooler.layers.1.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 820 |
+
"tokenizer.attention_pooler.layers.1.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 821 |
+
"tokenizer.attention_pooler.layers.1.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 822 |
+
"tokenizer.attention_pooler.layers.1.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
|
| 823 |
+
"tokenizer.attention_pooler.layers.1.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 824 |
+
"tokenizer.attention_pooler.layers.1.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 825 |
+
"tokenizer.attention_pooler.layers.1.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
|
| 826 |
+
"tokenizer.attention_pooler.layers.1.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 827 |
+
"tokenizer.attention_pooler.layers.1.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 828 |
+
"tokenizer.attention_pooler.norm.weight": "model-00004-of-00004.safetensors",
|
| 829 |
+
"tokenizer.attention_pooler.special_token": "model-00004-of-00004.safetensors",
|
| 830 |
+
"tokenizer.audio_acoustic_proj.bias": "model-00004-of-00004.safetensors",
|
| 831 |
+
"tokenizer.audio_acoustic_proj.weight": "model-00004-of-00004.safetensors",
|
| 832 |
+
"tokenizer.quantizer.project_in.bias": "model-00004-of-00004.safetensors",
|
| 833 |
+
"tokenizer.quantizer.project_in.weight": "model-00004-of-00004.safetensors",
|
| 834 |
+
"tokenizer.quantizer.project_out.bias": "model-00004-of-00004.safetensors",
|
| 835 |
+
"tokenizer.quantizer.project_out.weight": "model-00004-of-00004.safetensors"
|
| 836 |
+
}
|
| 837 |
+
}
|
modeling_acestep_v15_xl_turbo.py
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
silence_latent.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a778e9dd942f5e8b2c09c55370782d318834432b03dabbcdf70e6ed49ad6358b
|
| 3 |
+
size 3841215
|