CoolFace
Modelpublic

RichardErkhov/HuggingFaceH4_-_mistral-7b-grok-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
1likes336downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mistral-7b-grok - GGUF

  • —Model creator: https://huggingface.co/HuggingFaceH4/
  • —Original model: https://huggingface.co/HuggingFaceH4/mistral-7b-grok/
NameQuant methodSize
mistral-7b-grok.Q2_K.ggufQ2_K2.53GB
mistral-7b-grok.IQ3_XS.ggufIQ3_XS2.81GB
mistral-7b-grok.IQ3_S.ggufIQ3_S2.96GB
mistral-7b-grok.Q3_K_S.ggufQ3KS2.95GB
mistral-7b-grok.IQ3_M.ggufIQ3_M3.06GB
mistral-7b-grok.Q3_K.ggufQ3_K3.28GB
mistral-7b-grok.Q3_K_M.ggufQ3KM3.28GB
mistral-7b-grok.Q3_K_L.ggufQ3KL3.56GB
mistral-7b-grok.IQ4_XS.ggufIQ4_XS3.67GB
mistral-7b-grok.Q4_0.ggufQ4_03.83GB
mistral-7b-grok.IQ4_NL.ggufIQ4_NL3.87GB
mistral-7b-grok.Q4_K_S.ggufQ4KS3.86GB
mistral-7b-grok.Q4_K.ggufQ4_K4.07GB
mistral-7b-grok.Q4_K_M.ggufQ4KM4.07GB
mistral-7b-grok.Q4_1.ggufQ4_14.24GB
mistral-7b-grok.Q5_0.ggufQ5_04.65GB
mistral-7b-grok.Q5_K_S.ggufQ5KS4.65GB
mistral-7b-grok.Q5_K.ggufQ5_K4.78GB
mistral-7b-grok.Q5_K_M.ggufQ5KM4.78GB
mistral-7b-grok.Q5_1.ggufQ5_15.07GB
mistral-7b-grok.Q6_K.ggufQ6_K5.53GB

Original model description: --- license: apache-2.0 base_model: mistralai/Mistral-7B-v0.1 tags:

  • —alignment-handbook
  • —generatedfromtrainer datasets:
  • —HuggingFaceH4/grok-conversation-harmless
  • —HuggingFaceH4/ultrachat_200k model-index:
  • —name: mistral-7b-grok 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. -->

Mistral 7B Grok

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 that has been aligned via Constitutional AI to mimic the style of xAI's Grok assistant.

It achieves the following results on the evaluation set:

  • —Loss: 0.9348

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: 2e-05
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —distributed_type: multi-GPU
  • —num_devices: 8
  • —gradientaccumulationsteps: 4
  • —totaltrainbatch_size: 256
  • —totalevalbatch_size: 64
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_ratio: 0.1
  • —num_epochs: 1

Training results

Training LossEpochStepValidation Loss
0.93261.05450.9348

Framework versions

  • —Transformers 4.36.2
  • —Pytorch 2.1.2+cu121
  • —Datasets 2.16.1
  • —Tokenizers 0.15.0