jrahn's picture
Adding Evaluation Results (#1)
2edff46 verified
metadata
language:
  - en
license: llama3
library_name: peft
tags:
  - generated_from_trainer
base_model: meta-llama/Meta-Llama-3-8B-Instruct
datasets:
  - Norquinal/claude_multi_instruct_30k
model-index:
  - name: llama-3-8b-claudstruct-v3
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

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

load_in_8bit: false
load_in_4bit: true
strict: false

chat_template: llama3
datasets:
  - path: Norquinal/claude_multi_instruct_30k
    type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/llama-3-8b-claudstruct-v3/

adapter: qlora
lora_model_dir:

sequence_len: 512
sample_packing: false
pad_to_sequence_len: true

lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 8
num_epochs: 2
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.00001

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

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

warmup_steps: 10
evals_per_epoch: 4
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
special_tokens:
  pad_token: <|end_of_text|>

llama-3-8b-claudstruct-v3

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the Norquinal/claude_multi_instruct_30k dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6226

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: 8
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 2
  • total_train_batch_size: 16
  • total_eval_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 10
  • num_epochs: 2

Training results

Training Loss Epoch Step Validation Loss
2.2209 0.0007 1 2.0399
1.7842 0.2502 341 1.6960
1.6914 0.5004 682 1.6590
1.6757 0.7506 1023 1.6414
1.5182 1.0007 1364 1.6319
1.8421 1.2509 1705 1.6264
1.7271 1.5011 2046 1.6237
1.4817 1.7513 2387 1.6226

Framework versions

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

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 65.62
AI2 Reasoning Challenge (25-Shot) 58.96
HellaSwag (10-Shot) 80.05
MMLU (5-Shot) 64.55
TruthfulQA (0-shot) 51.76
Winogrande (5-shot) 74.19
GSM8k (5-shot) 64.22