See axolotl config

axolotl version: 0.4.0

base_model: meta-llama/Meta-Llama-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
hub_model_id:  KolaGang/Red_Llama_3_base
hub_strategy: end
load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: Drewskidang/chatlaw
    type: sharegpt
    conversation: chatml
  - path: Drewskidang/tool
    type: sharegpt
    conversation: chatml
  - path: digitalpipelines/samantha-1.1-uncensored
    type: sharegpt
    conversation: chatml
  - path: KolaGang/mergers
    type: alpaca
 
 



dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./out

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
eval_sample_packing: false

wandb_project: Legal_Llama 
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 4
num_epochs: 3
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 2e-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
flash_attn_cross_entropy: false
flash_attn_rms_norm: true
flash_attn_fuse_qkv: false
flash_attn_fuse_mlp: true
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-4

warmup_steps: 100
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed: deepspeed_configs/zero1.json # multi-gpu only
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|end_of_text|>

Red_Llama_3_base

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: 0.8173

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=0.0001
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
1.1958	0.03	1	1.1846
1.0515	0.49	19	1.0706
0.952	0.99	38	0.9385
0.9038	1.44	57	0.8796
0.8679	1.94	76	0.8469
0.7675	2.39	95	0.8280
0.7643	2.88	114	0.8173

Framework versions

• Transformers 4.40.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0