---
language:
- en
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- kejian/codeparrot-train-more-filter-3.3b-cleaned
model-index:
- name: kejian/final-awr
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# kejian/final-awr

This model was trained from scratch on the kejian/codeparrot-train-more-filter-3.3b-cleaned dataset.

## 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: 0.001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- training_steps: 12588
- mixed_precision_training: Native AMP

### Framework versions

- Transformers 4.23.0
- Pytorch 1.13.0+cu116
- Datasets 2.0.0
- Tokenizers 0.12.1


# Full config
{'dataset': {'datasets': ['kejian/codeparrot-train-more-filter-3.3b-cleaned'],
             'is_split_by_sentences': True},
 'generation': {'batch_size': 64,
                'metrics_configs': [{}, {'n': 1}, {}],
                'scenario_configs': [{'display_as_html': True,
                                      'generate_kwargs': {'do_sample': True,
                                                          'eos_token_id': 0,
                                                          'max_length': 640,
                                                          'min_length': 10,
                                                          'temperature': 0.7,
                                                          'top_k': 0,
                                                          'top_p': 0.9},
                                      'name': 'unconditional',
                                      'num_samples': 512},
                                     {'display_as_html': True,
                                      'generate_kwargs': {'do_sample': True,
                                                          'eos_token_id': 0,
                                                          'max_length': 272,
                                                          'min_length': 10,
                                                          'temperature': 0.7,
                                                          'top_k': 0,
                                                          'top_p': 0.9},
                                      'name': 'functions',
                                      'num_samples': 512,
                                      'prompts_path': 'resources/functions_csnet.jsonl',
                                      'use_prompt_for_scoring': True}],
                'scorer_config': {}},
 'kl_gpt3_callback': {'gpt3_kwargs': {'model_name': 'code-cushman-001'},
                      'max_tokens': 64,
                      'num_samples': 4096},
 'model': {'from_scratch': True,
           'gpt2_config_kwargs': {'reorder_and_upcast_attn': True,
                                  'scale_attn_by': True},
           'model_kwargs': {'value_head_config': {'is_detached': False}},
           'path_or_name': 'codeparrot/codeparrot-small'},
 'objective': {'alpha': 0.05, 'beta': 1, 'name': 'AWR'},
 'tokenizer': {'path_or_name': 'codeparrot/codeparrot-small'},
 'training': {'dataloader_num_workers': 0,
              'effective_batch_size': 256,
              'evaluation_strategy': 'no',
              'fp16': True,
              'hub_model_id': 'kejian/final-awr',
              'hub_strategy': 'all_checkpoints',
              'learning_rate': 0.001,
              'logging_first_step': True,
              'logging_steps': 1,
              'num_tokens': 3300000000.0,
              'output_dir': 'training_output',
              'per_device_eval_batch_size': 16,
              'per_device_train_batch_size': 16,
              'push_to_hub': True,
              'remove_unused_columns': False,
              'save_steps': 5000,
              'save_strategy': 'steps',
              'seed': 42,
              'warmup_ratio': 0.01,
              'weight_decay': 0.1}}

# Wandb URL:
https://wandb.ai/kejian/uncategorized/runs/3a8ddwg4