See axolotl config

axolotl version: 0.10.0

base_model: meta-llama/Llama-3.1-8B-Instruct
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false

# 1:1 mixing: 23k AR docs + 23k DCLM docs (~46k total)
# Axolotl concatenates and shuffles both datasets
datasets:
  - path: cfierro/simpleqa_wiki_ar_Llama-3.1-8B-Instruct
    type: completion
    field: text
    split: test
  - path: cfierro/dclm_baseline_sampled
    type: completion
    field: text
    split: sampled_23k
dataset_prepared_path: /scratch/project/eu-25-39/knowledge-ft/axolotl/datasets/llama-8b/simpleqa-ar-dclm-1to1
val_set_size: 0.0
output_dir: /scratch/project/eu-25-39/knowledge-ft/axolotl/models/llama-3.1-8b-fft-simpleqa-ar-dclm-1to1
hub_model_id: llama-3.1-8b-fft-simpleqa-ar-dclm-1to1

sequence_len: 4096
sample_packing: true
eval_sample_packing: false

# No LoRA — full fine-tuning

wandb_project: knowledge-ft
wandb_entity: cfierro
wandb_watch:
wandb_name: llama-3.1-8b-fft-simpleqa-ar-dclm-1to1
wandb_log_model: "false"

# Multi-GPU settings
# micro_batch=1 to fit full FT in memory (ZeRO-3 on 4x A100 40GB)
# grad_accum=2 to keep same effective batch size (1 * 2 * 4 GPUs = 8)
# ~46k examples with sample_packing (~6-8 per packed seq) → ~6.5k steps for 1 pass
# Using 6000 steps same as baseline to keep step count comparable
gradient_accumulation_steps: 2
micro_batch_size: 1
max_steps: 6000

optimizer: adamw_bnb_8bit
lr_scheduler: constant
learning_rate: 1e-5

bf16: auto
tf32: false

gradient_checkpointing: true
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.03
save_steps: 1000
save_total_limit: 1
load_best_model_at_end: true
weight_decay: 0.0
special_tokens:
   pad_token: <|end_of_text|>

# DeepSpeed ZeRO Stage 3 - shards model weights, gradients, and optimizer across GPUs
deepspeed: deepspeed_configs/zero3.json

llama-3.1-8b-fft-simpleqa-ar-dclm-1to1

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the cfierro/simpleqa_wiki_ar_Llama-3.1-8B-Instruct and the cfierro/dclm_baseline_sampled datasets.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• lr_scheduler_warmup_steps: 37
• training_steps: 6000

Training results

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu128
• Datasets 3.5.0
• Tokenizers 0.22.2