GaetanMichelet/Mistral-7B_task-2_180-samples_config-2

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Mistral-7Btask-2180-samples_config-2

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60fttask-2, the GaetanMichelet/chat-120fttask-2 and the GaetanMichelet/chat-180fttask-2 datasets. It achieves the following results on the evaluation set:

• —Loss: 0.5403

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
• —trainbatchsize: 1
• —evalbatchsize: 1
• —seed: 42
• —distributed_type: multi-GPU
• —gradientaccumulationsteps: 16
• —totaltrainbatch_size: 16
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 50

Training results

Framework versions

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