murugeshmarvel/QAD-llama3.1-8b-iter4set4

See axolotl config

axolotl version: 0.5.2

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

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: murugeshmarvel/iter4_qad_set4
    type: alpaca:instruct
    train_on_split: train

test_datasets:
  - path: murugeshmarvel/iter4_qad_set4
    type: alpaca:instruct
    split: test

unfrozen_parameters:
- ^lm_head.weight$
- ^model.embed_tokens.weight$
# input_layernorm layers
- model.layers.0.input_layernorm
- model.layers.1.input_layernorm
- model.layers.2.input_layernorm
- model.layers.3.input_layernorm
- model.layers.4.input_layernorm
- model.layers.5.input_layernorm
- model.layers.6.input_layernorm
- model.layers.7.input_layernorm
- model.layers.8.input_layernorm
- model.layers.9.input_layernorm
- model.layers.10.input_layernorm
- model.layers.11.input_layernorm
- model.layers.12.input_layernorm
- model.layers.13.input_layernorm
- model.layers.14.input_layernorm
- model.layers.15.input_layernorm
# lm_head layers
# mlp.down_proj layers
- model.layers.0.mlp.down_proj
- model.layers.1.mlp.down_proj
- model.layers.30.mlp.down_proj
- model.layers.2.mlp.down_proj
- model.layers.21.mlp.down_proj
- model.layers.29.mlp.down_proj
- model.layers.22.mlp.down_proj
- model.layers.5.mlp.down_proj
- model.layers.4.mlp.down_proj
- model.layers.20.mlp.down_proj
- model.layers.23.mlp.down_proj
- model.layers.19.mlp.down_proj
- model.layers.3.mlp.down_proj
- model.layers.17.mlp.down_proj
- model.layers.6.mlp.down_proj
- model.layers.31.mlp.down_proj
# mlp.gate_proj layers
- model.layers.1.mlp.gate_proj
- model.layers.2.mlp.gate_proj
- model.layers.3.mlp.gate_proj
- model.layers.4.mlp.gate_proj
- model.layers.0.mlp.gate_proj
- model.layers.25.mlp.gate_proj
- model.layers.26.mlp.gate_proj
- model.layers.5.mlp.gate_proj
- model.layers.24.mlp.gate_proj
- model.layers.28.mlp.gate_proj
- model.layers.23.mlp.gate_proj
- model.layers.27.mlp.gate_proj
- model.layers.21.mlp.gate_proj
- model.layers.22.mlp.gate_proj
- model.layers.29.mlp.gate_proj
- model.layers.20.mlp.gate_proj
# mlp.up_proj layers
- model.layers.4.mlp.up_proj
- model.layers.3.mlp.up_proj
- model.layers.0.mlp.up_proj
- model.layers.7.mlp.up_proj
- model.layers.5.mlp.up_proj
- model.layers.6.mlp.up_proj
- model.layers.2.mlp.up_proj
- model.layers.1.mlp.up_proj
- model.layers.8.mlp.up_proj
- model.layers.14.mlp.up_proj
- model.layers.12.mlp.up_proj
- model.layers.9.mlp.up_proj
- model.layers.15.mlp.up_proj
- model.layers.17.mlp.up_proj
- model.layers.13.mlp.up_proj
- model.layers.19.mlp.up_proj
# model.embed_tokens layers
# model.norm layers
# post_attention_layernorm layers
- model.layers.0.post_attention_layernorm
- model.layers.1.post_attention_layernorm
- model.layers.2.post_attention_layernorm
- model.layers.3.post_attention_layernorm
- model.layers.4.post_attention_layernorm
- model.layers.5.post_attention_layernorm
- model.layers.6.post_attention_layernorm
- model.layers.7.post_attention_layernorm
- model.layers.8.post_attention_layernorm
- model.layers.9.post_attention_layernorm
- model.layers.10.post_attention_layernorm
- model.layers.11.post_attention_layernorm
- model.layers.12.post_attention_layernorm
- model.layers.13.post_attention_layernorm
- model.layers.14.post_attention_layernorm
- model.layers.15.post_attention_layernorm
# self_attn.k_proj layers
- model.layers.29.self_attn.k_proj
- model.layers.25.self_attn.k_proj
- model.layers.23.self_attn.k_proj
- model.layers.28.self_attn.k_proj
- model.layers.21.self_attn.k_proj
- model.layers.19.self_attn.k_proj
- model.layers.22.self_attn.k_proj
- model.layers.20.self_attn.k_proj
- model.layers.24.self_attn.k_proj
- model.layers.31.self_attn.k_proj
- model.layers.27.self_attn.k_proj
- model.layers.26.self_attn.k_proj
- model.layers.17.self_attn.k_proj
- model.layers.11.self_attn.k_proj
- model.layers.18.self_attn.k_proj
- model.layers.14.self_attn.k_proj
# self_attn.o_proj layers
- model.layers.14.self_attn.o_proj
- model.layers.7.self_attn.o_proj
- model.layers.5.self_attn.o_proj
- model.layers.11.self_attn.o_proj
- model.layers.6.self_attn.o_proj
- model.layers.24.self_attn.o_proj
- model.layers.9.self_attn.o_proj
- model.layers.13.self_attn.o_proj
- model.layers.10.self_attn.o_proj
- model.layers.12.self_attn.o_proj
- model.layers.8.self_attn.o_proj
- model.layers.25.self_attn.o_proj
- model.layers.21.self_attn.o_proj
- model.layers.23.self_attn.o_proj
- model.layers.15.self_attn.o_proj
- model.layers.16.self_attn.o_proj
# self_attn.q_proj layers
- model.layers.8.self_attn.q_proj
- model.layers.13.self_attn.q_proj
- model.layers.9.self_attn.q_proj
- model.layers.14.self_attn.q_proj
- model.layers.10.self_attn.q_proj
- model.layers.11.self_attn.q_proj
- model.layers.0.self_attn.q_proj
- model.layers.15.self_attn.q_proj
- model.layers.1.self_attn.q_proj
- model.layers.6.self_attn.q_proj
- model.layers.5.self_attn.q_proj
- model.layers.7.self_attn.q_proj
- model.layers.12.self_attn.q_proj
- model.layers.16.self_attn.q_proj
- model.layers.17.self_attn.q_proj
- model.layers.26.self_attn.q_proj
# self_attn.v_proj layers
- model.layers.26.self_attn.v_proj
- model.layers.17.self_attn.v_proj
- model.layers.3.self_attn.v_proj
- model.layers.28.self_attn.v_proj
- model.layers.29.self_attn.v_proj
- model.layers.21.self_attn.v_proj
- model.layers.15.self_attn.v_proj
- model.layers.16.self_attn.v_proj
- model.layers.20.self_attn.v_proj
- model.layers.25.self_attn.v_proj
- model.layers.6.self_attn.v_proj
- model.layers.23.self_attn.v_proj
- model.layers.4.self_attn.v_proj
- model.layers.1.self_attn.v_proj
- model.layers.14.self_attn.v_proj
- model.layers.22.self_attn.v_proj

dataset_prepared_path: last_run_prepared
output_dir: ./outputs/qad_base_fft_out

sequence_len: 8192  
sample_packing: true
pad_to_sequence_len: true

wandb_project: QAD_ITER4_SET4
wandb_entity:
wandb_watch:
wandb_name: QAD_ITER4_SET4_4GPU
wandb_log_model:

gradient_accumulation_steps: 8
micro_batch_size: 2
eval_batch_size: 3
num_epochs: 5
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 1e-5

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
s2_attention:

warmup_steps: 10
evals_per_epoch: 2
eval_table_size:
eval_max_new_tokens: 4096
saves_per_epoch: 4
save_total_limit: 2
debug:
deepspeed: 
weight_decay: 0.05
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|end_of_text|>

outputs/qad_base_fft_out

This model is a fine-tuned version of meta-llama/Llama-3.1-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1032

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: 2
• eval_batch_size: 3
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• total_eval_batch_size: 6
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
0.1159	0.5018	102	0.1200
0.1883	1.0031	204	0.1116
0.086	1.5049	306	0.1093
0.1044	2.0080	408	0.1060
0.0909	2.5098	510	0.1051
0.0754	3.0123	612	0.1036
0.0697	3.5141	714	0.1034
0.0794	4.0160	816	0.1032
0.0839	4.5178	918	0.1032

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3