Hugging Face's logo

ardaspear
/
a15cb6da-e674-42f9-8be4-692b0f7970da

PEFT
Safetensors
llama
axolotl
Generated from Trainer
custom_code
Model card Files Community
a15cb6da-e674-42f9-8be4-692b0f7970da / README.md
ardaspear's picture
ardaspear
End of training
25093dc verified
preview code
|
raw
history blame contribute delete
4.34 kB
metadata
library_name: peft
base_model: NousResearch/CodeLlama-13b-hf-flash
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: a15cb6da-e674-42f9-8be4-692b0f7970da
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/CodeLlama-13b-hf-flash
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - bc1633ec07335911_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/bc1633ec07335911_train_data.json
  type:
    field_input: text
    field_instruction: hypothesis
    field_output: label_text
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: ardaspear/a15cb6da-e674-42f9-8be4-692b0f7970da
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 72GB
max_steps: 100
micro_batch_size: 4
mlflow_experiment_name: /tmp/bc1633ec07335911_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: leixa-personal
wandb_mode: online
wandb_name: a15cb6da-e674-42f9-8be4-692b0f7970da
wandb_project: Gradients-On-Two
wandb_run: your_name
wandb_runid: a15cb6da-e674-42f9-8be4-692b0f7970da
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

a15cb6da-e674-42f9-8be4-692b0f7970da

This model is a fine-tuned version of NousResearch/CodeLlama-13b-hf-flash on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0000

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: 0.0001
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • optimizer: Use OptimizerNames.ADAMW_BNB 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: 10
  • training_steps: 100

Training results

Training Loss Epoch Step Validation Loss
No log 0.0088 1 7.8507
15.0507 0.0796 9 1.8900
0.0137 0.1593 18 0.0003
0.0001 0.2389 27 0.0000
0.0023 0.3186 36 0.0000
0.0001 0.3982 45 0.0000
0.0 0.4779 54 0.0000
0.0 0.5575 63 0.0000
0.0 0.6372 72 0.0000
0.0 0.7168 81 0.0000
0.0 0.7965 90 0.0000
0.0 0.8761 99 0.0000

Framework versions

  • PEFT 0.13.2
  • Transformers 4.46.0
  • Pytorch 2.5.0+cu124
  • Datasets 3.0.1
  • Tokenizers 0.20.1