| { |
| "_name_or_path": "mplus-8b", |
| "add_bos_embedding": true, |
| "add_decoder_lora": true, |
| "add_encoder_retriever": false, |
| "add_selector": true, |
| "architectures": [ |
| "MPlus" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 128000, |
| "decay_frequency": 1, |
| "detach_hidden_state": true, |
| "drop_memory_per_layer": true, |
| "eos_token_id": 128001, |
| "fix_poe_for_encoder": false, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 4096, |
| "initial_rf_when_moving_stm_to_ltm": 10, |
| "initializer_range": 0.02, |
| "intermediate_size": 14336, |
| "lora_config": { |
| "inference_mode": false, |
| "lora_alpha": 32, |
| "lora_dropout": 0.1, |
| "r": 16, |
| "target_modules": [ |
| "q_proj", |
| "v_proj", |
| "k_proj", |
| "up_proj", |
| "down_proj", |
| "gate_proj" |
| ] |
| }, |
| "ltm_initial_size": 2560, |
| "max_position_embeddings": 131072, |
| "max_seq_length": 2048, |
| "min_num_tokens": 25, |
| "mlp_bias": false, |
| "model_type": "llama", |
| "num_attention_heads": 32, |
| "num_blocks": 50, |
| "num_hidden_layers": 32, |
| "num_key_value_heads": 8, |
| "num_ltm_blocks": 10, |
| "num_memory_tokens": 12800, |
| "num_selector_layers": 2, |
| "num_tokens": 256, |
| "pad_token_id": 128001, |
| "pretraining_tp": 1, |
| "put_cached_dropped_memory_on_cpu": true, |
| "rms_norm_eps": 1e-05, |
| "rope_scaling": { |
| "factor": 8.0, |
| "high_freq_factor": 4.0, |
| "low_freq_factor": 1.0, |
| "original_max_position_embeddings": 8192, |
| "rope_type": "llama3" |
| }, |
| "rope_theta": 500000.0, |
| "selector_hidden_dim": 256, |
| "shrink_to_one_embedding": true, |
| "spread_embeddings": false, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float32", |
| "transformers_version": "4.48.2", |
| "update_ltm_frequency": 60, |
| "update_ltm_mode": "immediate", |
| "use_cache": true, |
| "vocab_size": 128256, |
| "wrap_memory": false |
| } |
|
|
