GaetanMichelet's picture
End of training
15693a7 verified
metadata
base_model: mistralai/Mistral-7B-Instruct-v0.3
datasets:
  - GaetanMichelet/chat-60_ft_task-2_auto
  - GaetanMichelet/chat-120_ft_task-2_auto
library_name: peft
license: apache-2.0
tags:
  - alignment-handbook
  - trl
  - sft
  - generated_from_trainer
model-index:
  - name: Mistral-7B_task-2_120-samples_config-1_full_auto
    results: []

Mistral-7B_task-2_120-samples_config-1_full_auto

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60_ft_task-2_auto and the GaetanMichelet/chat-120_ft_task-2_auto datasets. It achieves the following results on the evaluation set:

  • Loss: 0.7671

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: 0.0001
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss
1.067 1.0 11 1.0423
0.8773 2.0 22 0.9271
0.781 3.0 33 0.8065
0.79 4.0 44 0.7790
0.7295 5.0 55 0.7671
0.6803 6.0 66 0.7684
0.6231 7.0 77 0.7836
0.5575 8.0 88 0.8163
0.452 9.0 99 0.8697
0.3535 10.0 110 0.9049
0.3253 11.0 121 0.9732
0.2393 12.0 132 1.0470

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.0
  • Pytorch 2.1.2+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1