pbevan11's picture
Training in progress, step 30
bd54982 verified
|
raw
history blame
3.47 kB
metadata
base_model: meta-llama/Meta-Llama-3.1-8B
library_name: peft
license: llama3.1
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: llama-3.1-8b-ocr-correction
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

base_model: meta-llama/Meta-Llama-3.1-8B
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

lora_fan_in_fan_out: false
data_seed: 49
seed: 49

datasets:
  - path: ft_data/alpaca_data.jsonl
    type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.1
output_dir: ./qlora-alpaca-out
hub_model_id: pbevan11/llama-3.1-8b-ocr-correction

adapter: qlora
lora_model_dir:

sequence_len: 8192
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true

lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:
  - gate_proj
  - down_proj
  - up_proj
  - q_proj
  - v_proj
  - k_proj
  - o_proj

wandb_project: ocr-ft
wandb_entity: sncds
wandb_name: llama31

gradient_accumulation_steps: 4
micro_batch_size: 2 # was 16
eval_batch_size: 2 # was 16
num_epochs: 2
optimizer: paged_adamw_32bit
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

loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3

warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  pad_token: "<|end_of_text|>"

Visualize in Weights & Biases

llama-3.1-8b-ocr-correction

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6569

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: 49
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 8
  • 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
0.6009 0.8 1 0.6584
0.5865 1.2 2 0.6569

Framework versions

  • PEFT 0.11.1
  • Transformers 4.43.2
  • Pytorch 2.1.2+cu118
  • Datasets 2.19.1
  • Tokenizers 0.19.1