RichardErkhov/terry69_-_preference_p0.2_seed42_level2_raremixbatch16-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

preferencep0.2seed42level2raremixbatch16 - GGUF

• —Model creator: https://huggingface.co/terry69/
• —Original model: https://huggingface.co/terry69/preferencep0.2seed42level2raremixbatch16/

Original model description: --- libraryname: transformers license: apache-2.0 basemodel: mistralai/Mistral-7B-Instruct-v0.2 tags:

• —alignment-handbook
• —trl
• —sft
• —generatedfromtrainer
• —trl
• —sft
• —alignment-handbook
• —generatedfromtrainer datasets:
• —preference-data model-index:
• —name: preferencep0.2seed42level2raremixbatch16 results: [] ---

<!-- 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. -->

preferencep0.2seed42level2raremixbatch16

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the preference-data dataset. It achieves the following results on the evaluation set:

• —Loss: 0.2978

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-05
• —trainbatchsize: 4
• —evalbatchsize: 1
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 4
• —totaltrainbatch_size: 16
• —totalevalbatch_size: 4
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 1

Training results

Framework versions

• —Transformers 4.44.2
• —Pytorch 2.3.1+cu121
• —Datasets 2.19.1
• —Tokenizers 0.19.1