NekoMikoReimu's picture
Upload folder using huggingface_hub
14e8772
metadata
license: apache-2.0
base_model: falche/opennovel_oc2_01a_7b
tags:
  - generated_from_trainer
model-index:
  - name: out
    results: []

Built with Axolotl

out

This model is a fine-tuned version of falche/opennovel_oc2_01a_7b on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.3955

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: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 3
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 6
  • total_eval_batch_size: 3
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss
3.1649 0.0 1 3.3906
3.6247 0.05 528 3.7267
3.5584 0.1 1056 3.7235
3.4741 0.15 1584 3.7208
3.4915 0.2 2112 3.6959
3.4849 0.25 2640 3.6844
3.4466 0.3 3168 3.6704
3.2705 0.35 3696 3.6390
3.2936 0.4 4224 3.6249
3.4238 0.45 4752 3.6018
3.3485 0.5 5280 3.5783
3.2596 0.55 5808 3.5661
3.2277 0.6 6336 3.5371
3.2635 0.65 6864 3.5157
3.1735 0.7 7392 3.4962
3.1552 0.75 7920 3.4712
3.0954 0.8 8448 3.4501
3.09 0.85 8976 3.4298
3.1094 0.9 9504 3.4096
2.9722 0.95 10032 3.3955

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1