README.md

metadata

license: apache-2.0
base_model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
tags:
  - trl
  - sft
  - generated_from_trainer
datasets:
  - generator
model-index:
  - name: leaderboard_inst_v1.5_dedup-eeve_EEVE-Korean-Instruct-10.8B-v1.0_SFT
    results: []

leaderboard_inst_v1.5_dedup-eeve_EEVE-Korean-Instruct-10.8B-v1.0_SFT

This model is a fine-tuned version of yanolja/EEVE-Korean-Instruct-10.8B-v1.0 on the generator 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: 1e-06
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 3

Training results

Framework versions

• Transformers 4.38.1
• Pytorch 2.2.1+cu118
• Datasets 2.17.1
• Tokenizers 0.15.2