metadata

library_name: transformers
license: llama3.1
base_model: meta-llama/Meta-Llama-3.1-8B
tags:
  - axolotl
  - generated_from_trainer
datasets:
  - penfever/allenai_WildChat-1M-Full-Qwen_Qwen2.5-72B-Instruct
  - HuggingFaceH4/ifeval-like-data
  - penfever/mmlu-it
  - penfever/OpenMathInstruct-1-alpaca-chat
model-index:
  - name: Llama-3-8B-WildChat-500k-qwen2-72b-lbt
    results: []

See axolotl config

axolotl version: 0.6.0

base_model: meta-llama/Meta-Llama-3.1-8B

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_fused_linear_cross_entropy: true

strict: false

chat_template: llama3
datasets:
  - path: penfever/allenai_WildChat-1M-Full-Qwen_Qwen2.5-72B-Instruct
    type: chat_template
    split: train[:50%]
    field_messages: conversation
    message_field_role: role
    message_field_content: content
  - path: HuggingFaceH4/ifeval-like-data
    type: chat_template
    split: train
    field_messages: messages
    message_field_role: role
    message_field_content: content
  - path: penfever/mmlu-it
    type: jeopardy
    split: auxiliary_train
  - path: penfever/OpenMathInstruct-1-alpaca-chat
    type: alpaca_chat.load_qa
    split: train[:10%]

dataset_prepared_path: /scratch/bf996/axolotl/datasets/wildchat-500k-qwen2-72b-lbt
val_set_size: 0.02
output_dir: /scratch/bf996/axolotl/outputs/llama-3-8b-wildchat-500k-qwen2-72b-lbt

sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true

wandb_project: lm-evals
wandb_entity:
wandb_watch:
wandb_name: Llama-3-8B-WildChat-qwen2-72b-lbt
wandb_log_model:
hub_model_id: penfever/Llama-3-8B-WildChat-500k-qwen2-72b-lbt


gradient_accumulation_steps: 8
micro_batch_size: 2
num_epochs: 1
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 2e-5

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

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 100
evals_per_epoch: 0
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdp_limit_all_gathers: true
  fsdp_sync_module_states: true
  fsdp_offload_params: true
  fsdp_use_orig_params: false
  fsdp_cpu_ram_efficient_loading: true
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
  fsdp_state_dict_type: FULL_STATE_DICT
  fsdp_sharding_strategy: FULL_SHARD
  fsdp_backward_prefetch: BACKWARD_PRE
special_tokens:
  pad_token: <|finetune_right_pad_id|>
  eos_token: <|eot_id|>

Llama-3-8B-WildChat-500k-qwen2-72b-lbt

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the penfever/allenai_WildChat-1M-Full-Qwen_Qwen2.5-72B-Instruct, the HuggingFaceH4/ifeval-like-data, the penfever/mmlu-it and the penfever/OpenMathInstruct-1-alpaca-chat datasets. It achieves the following results on the evaluation set:

Loss: 0.6168

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: 2
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 8
total_train_batch_size: 64
total_eval_batch_size: 8
optimizer: Use adamw_torch 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: 100
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.631	1.0	1912	0.6168

Framework versions

Transformers 4.48.0.dev0
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.21.0