See axolotl config

axolotl version: 0.4.1

base_model: meta-llama/Meta-Llama-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

data_seed: 42
seed: 42

datasets:
  - path: data/isaf_press_releases_ft.jsonl
    conversation: alpaca
    type: sharegpt
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./outputs/llama3/lora-out
hub_model_id: strickvl/isafpr-llama3-lora

sequence_len: 2048
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true

adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_modules_to_save:
  - embed_tokens
  - lm_head

wandb_project: isaf_pr_ft
wandb_entity: strickvl
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

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: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
   pad_token: <|end_of_text|>

isafpr-llama3-lora

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0371

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: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
2.0023	0.0173	1	2.0120
0.0975	0.2597	15	0.0792
0.0576	0.5195	30	0.0586
0.0317	0.7792	45	0.0476
0.0367	1.0390	60	0.0445
0.0315	1.2078	75	0.0421
0.0249	1.4675	90	0.0429
0.0302	1.7273	105	0.0380
0.0264	1.9870	120	0.0376
0.0184	2.1515	135	0.0362
0.0174	2.4113	150	0.0366
0.0152	2.6710	165	0.0373
0.016	2.9307	180	0.0361
0.0128	3.0996	195	0.0361
0.0172	3.3593	210	0.0369
0.0086	3.6190	225	0.0371

Framework versions

• PEFT 0.11.1
• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1