SystemAdmin123's picture
Training in progress, step 320
ceaffa1 verified
adapter: lora
base_model: NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer
batch_size: 64
bf16: true
chat_template: tokenizer_default_fallback_alpaca
datasets:
- format: custom
path: https://gradients.s3.eu-north-1.amazonaws.com/1ae63171fbf66651_train_data.json?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAVVZOOA7SA4UOFLPI%2F20250210%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Date=20250210T120931Z&X-Amz-Expires=604800&X-Amz-SignedHeaders=host&X-Amz-Signature=bbe3efc2d3663c793088c33c43c09e6fbc67e548bccdd8400f529787d50520a6
type:
field_input: headline_a
field_instruction: rendered_input
field_output: headline_b
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
device_map: auto
eval_sample_packing: false
eval_steps: 0.1
flash_attention: true
gradient_checkpointing: true
group_by_length: true
hub_model_id: SystemAdmin123/ce7a539a-b639-41f4-8114-4dc65059452a
hub_strategy: checkpoint
learning_rate: 0.0001
logging_steps: 10
lora_alpha: 256
lora_dropout: 0.1
lora_r: 128
lora_target_linear: true
lr_scheduler: cosine
max_steps: 8000
micro_batch_size: 1
model_type: AutoModelForCausalLM
num_epochs: 10000
optimizer: adamw_bnb_8bit
output_dir: /root/.sn56/axolotl/tmp/ce7a539a-b639-41f4-8114-4dc65059452a
pad_to_sequence_len: true
resize_token_embeddings_to_32x: false
sample_packing: true
save_steps: 40
save_total_limit: 1
sequence_len: 2048
tokenizer_type: PreTrainedTokenizerFast
torch_dtype: bf16
training_args_kwargs:
disable_tqdm: true
hub_private_repo: true
save_only_model: true
trust_remote_code: true
val_set_size: 0.01
wandb_entity: ''
wandb_mode: online
wandb_name: NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer-https://gradients.s3.eu-north-1.amazonaws.com/1ae63171fbf66651_train_data.json?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAVVZOOA7SA4UOFLPI%2F20250210%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Date=20250210T120931Z&X-Amz-Expires=604800&X-Amz-SignedHeaders=host&X-Amz-Signature=bbe3efc2d3663c793088c33c43c09e6fbc67e548bccdd8400f529787d50520a6
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: default
warmup_ratio: 0.05