metadata
tags:
- generated_from_trainer
model-index:
- name: flashatt1_llama213B_yyj_deruzh.shot3.highcomet_-1.0
results: []
flashatt1_llama213B_yyj_deruzh.shot3.highcomet_-1.0
This model is a fine-tuned version of /mnt/nanjing3cephfs/pr-others-prctrans/lemonzeng/search/mGPT/models/llama-2-13b on the None 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: 2e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 1.0
Training results
Framework versions
- Transformers 4.28.0
- Pytorch 1.13.1
- Datasets 2.14.0
- Tokenizers 0.13.3