RichardErkhov/simonycl_-_llama-3-8b-instruct-single-judge-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama-3-8b-instruct-single-judge - GGUF

• —Model creator: https://huggingface.co/simonycl/
• —Original model: https://huggingface.co/simonycl/llama-3-8b-instruct-single-judge/

Original model description: --- libraryname: transformers license: llama3 basemodel: meta-llama/Meta-Llama-3-8B-Instruct tags:

• —alignment-handbook
• —generatedfromtrainer datasets:
• —simonycl/Meta-Llama-3-8B-Instructultrafeedbacksingle_judge model-index:
• —name: llama-3-8b-instruct-single-judge 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. -->

llama-3-8b-instruct-single-judge

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the simonycl/Meta-Llama-3-8B-Instructultrafeedbacksingle_judge dataset. It achieves the following results on the evaluation set:

• —Loss: 0.6593
• —Rewards/chosen: -1.3185
• —Rewards/rejected: -1.5107
• —Rewards/accuracies: 0.5935
• —Rewards/margins: 0.1922
• —Logps/rejected: -301.4192
• —Logps/chosen: -283.3882
• —Logits/rejected: -1.3226
• —Logits/chosen: -1.3593

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: 5e-07
• —trainbatchsize: 1
• —evalbatchsize: 2
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 4
• —gradientaccumulationsteps: 32
• —totaltrainbatch_size: 128
• —totalevalbatch_size: 8
• —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 2.4.0+cu121
• —Datasets 2.21.0
• —Tokenizers 0.19.1