Hugging Face's logo

Howard881010
/
climate-5day

PEFT
Safetensors
llama
Generated from Trainer
Model card Files Community
climate-5day / README.md
howard
update
9c90ebb
preview code
|
raw
history blame contribute delete
3.17 kB
metadata
base_model: unsloth/Meta-Llama-3.1-8B-Instruct
library_name: peft
license: llama3.1
tags:
  - generated_from_trainer
model-index:
  - name: finetune/output/climate-5day
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

base_model: unsloth/Meta-Llama-3.1-8B-Instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

chat_template: chatml
datasets:
  - path: Howard881010/climate-5day
    type: alpaca
    train_on_split: train
dataset_prepared_path:
output_dir: ./finetune/output/climate-5day

test_datasets:
  - path: Howard881010/climate-5day
    split: valid
    type: alpaca

adapter: lora
lora_model_dir:

sequence_len: 3200
sample_packing: false
pad_to_sequence_len: true

lora_r: 8
lora_alpha: 32
lora_dropout: 0.1
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project: finetune
wandb_entity:
wandb_watch:
wandb_name: climate-5day
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 4
num_epochs: 1
optimizer: adamw_hf
learning_rate: 0.00002
max_grad_norm: 1.0

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

gradient_checkpointing: true
logging_steps: 1
xformers_attention: 
flash_attention: true
eval_sample_packing: False

warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
weight_decay: 0.0
seed: 0
special_tokens:
  pad_token: "<|end_of_text|>"

finetune/output/climate-5day

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

  • Loss: 1.0853

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
  • total_train_batch_size: 32
  • total_eval_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 10
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss
1.7314 0.0028 1 1.3110
1.5815 0.2514 91 1.1213
1.5008 0.5028 182 1.0918
1.3762 0.7541 273 1.0853

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.0
  • Pytorch 2.4.0+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1