metadata

base_model: /apdcephfs_cq10_1567347/share_1567347/share_info/llm_models/zephyr-7b-sft-full
tags:
  - alignment-handbook
  - generated_from_trainer
datasets:
  - >-
    /apdcephfs_cq10_1567347/share_1567347/mingxiaoli/tmp_data/ultrafeedback_binarized
model-index:
  - name: zephyr-7b-Database-SimPO+_bs2ga8_again
    results: []

zephyr-7b-Database-SimPO+_bs2ga8_again

This model is a fine-tuned version of /apdcephfs_cq10_1567347/share_1567347/share_info/llm_models/zephyr-7b-sft-full on the /apdcephfs_cq10_1567347/share_1567347/mingxiaoli/tmp_data/ultrafeedback_binarized 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: 5e-07
train_batch_size: 2
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 128
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Framework versions

Transformers 4.41.0
Pytorch 2.3.0+cu118
Datasets 2.20.0
Tokenizers 0.19.1