Hugging Face's logo

vmpsergio
/
2bf69ce0-f510-4031-ad51-8df4fbb0e47d

PEFT
Safetensors
llama
axolotl
Generated from Trainer
Model card Files Community
2bf69ce0-f510-4031-ad51-8df4fbb0e47d / README.md
vmpsergio's picture
vmpsergio
End of training
e610cd3 verified
preview code
|
raw
history blame
4.09 kB
metadata
library_name: peft
license: other
base_model: NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: 2bf69ce0-f510-4031-ad51-8df4fbb0e47d
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - e9523ca25a42aaff_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/e9523ca25a42aaff_train_data.json
  type:
    field_input: endings
    field_instruction: ctx
    field_output: ctx_a
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device: cuda
early_stopping_patience: 1
eval_max_new_tokens: 128
eval_steps: 5
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: vmpsergio/2bf69ce0-f510-4031-ad51-8df4fbb0e47d
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 78GiB
max_steps: 30
micro_batch_size: 2
mlflow_experiment_name: /tmp/e9523ca25a42aaff_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
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: 10
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: true
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 6453abaa-2b4a-4c88-9a67-ea77743753fc
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 6453abaa-2b4a-4c88-9a67-ea77743753fc
warmup_steps: 10
weight_decay: 0.01
xformers_attention: true

2bf69ce0-f510-4031-ad51-8df4fbb0e47d

This model is a fine-tuned version of NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.1752

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: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 8
  • optimizer: Use OptimizerNames.ADAMW_TORCH 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: 30

Training results

Training Loss Epoch Step Validation Loss
No log 0.0004 1 1.5081
1.5825 0.0021 5 1.4666
1.3801 0.0042 10 1.3135
1.2339 0.0064 15 1.2224
1.1509 0.0085 20 1.1928
1.1488 0.0106 25 1.1780
1.0994 0.0127 30 1.1752

Framework versions

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