Hugging Face's logo

auxyus
/
81e8adf2-9dfa-4f49-bd89-57c64afb2de4

PEFT
Safetensors
llama
axolotl
Generated from Trainer
Model card Files Community
81e8adf2-9dfa-4f49-bd89-57c64afb2de4 / README.md
auxyus's picture
auxyus
End of training
d9863b9 verified
preview code
|
raw
history blame contribute delete
4.2 kB
metadata
library_name: peft
base_model: rayonlabs/83847950-33bc-4506-ba82-48653a06540a
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: 81e8adf2-9dfa-4f49-bd89-57c64afb2de4
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: rayonlabs/83847950-33bc-4506-ba82-48653a06540a
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - f4b7f90f2a0ae91e_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f4b7f90f2a0ae91e_train_data.json
  type:
    field_input: context
    field_instruction: question
    field_output: final_decision
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 5
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 150
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: false
group_by_length: false
hub_model_id: auxyus/81e8adf2-9dfa-4f49-bd89-57c64afb2de4
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 4.4e-05
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 10
lora_alpha: 32
lora_dropout: 0.2
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine_with_restarts
max_grad_norm: 1.0
max_memory:
  0: 75GB
max_steps: 600
micro_batch_size: 4
mlflow_experiment_name: /tmp/f4b7f90f2a0ae91e_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.999
  adam_epsilon: 1e-08
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 150
save_total_limit: 1
saves_per_epoch: null
sequence_len: 512
special_tokens:
  pad_token: <|end_of_text|>
strict: false
tf32: null
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: acopia-grant
wandb_mode: online
wandb_name: 53e001ba-f4ae-454f-a837-80c11f92d62b
wandb_project: Gradients-On-2
wandb_run: your_name
wandb_runid: 53e001ba-f4ae-454f-a837-80c11f92d62b
warmup_ratio: 0.1
weight_decay: 0.01
xformers_attention: null

81e8adf2-9dfa-4f49-bd89-57c64afb2de4

This model is a fine-tuned version of rayonlabs/83847950-33bc-4506-ba82-48653a06540a on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0464

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: 4.4e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.999,adam_epsilon=1e-08
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_steps: 60
  • training_steps: 600

Training results

Training Loss Epoch Step Validation Loss
No log 0.0000 1 2.0931
0.031 0.0030 150 0.0823
0.0013 0.0060 300 0.0778
0.0107 0.0090 450 0.0482
0.0833 0.0120 600 0.0464

Framework versions

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