CoolFace
Modelpublic

RichardErkhov/Hiranmai49_-_CFDistilGPT-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes43downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

CFDistilGPT - GGUF

  • —Model creator: https://huggingface.co/Hiranmai49/
  • —Original model: https://huggingface.co/Hiranmai49/CFDistilGPT/
NameQuant methodSize
CFDistilGPT.Q2_K.ggufQ2_K0.06GB
CFDistilGPT.IQ3_XS.ggufIQ3_XS0.07GB
CFDistilGPT.IQ3_S.ggufIQ3_S0.07GB
CFDistilGPT.Q3_K_S.ggufQ3KS0.07GB
CFDistilGPT.IQ3_M.ggufIQ3_M0.07GB
CFDistilGPT.Q3_K.ggufQ3_K0.07GB
CFDistilGPT.Q3_K_M.ggufQ3KM0.07GB
CFDistilGPT.Q3_K_L.ggufQ3KL0.07GB
CFDistilGPT.IQ4_XS.ggufIQ4_XS0.07GB
CFDistilGPT.Q4_0.ggufQ4_00.08GB
CFDistilGPT.IQ4_NL.ggufIQ4_NL0.08GB
CFDistilGPT.Q4_K_S.ggufQ4KS0.08GB
CFDistilGPT.Q4_K.ggufQ4_K0.08GB
CFDistilGPT.Q4_K_M.ggufQ4KM0.08GB
CFDistilGPT.Q4_1.ggufQ4_10.08GB
CFDistilGPT.Q5_0.ggufQ5_00.09GB
CFDistilGPT.Q5_K_S.ggufQ5KS0.09GB
CFDistilGPT.Q5_K.ggufQ5_K0.09GB
CFDistilGPT.Q5_K_M.ggufQ5KM0.09GB
CFDistilGPT.Q5_1.ggufQ5_10.09GB
CFDistilGPT.Q6_K.ggufQ6_K0.1GB
CFDistilGPT.Q8_0.ggufQ8_00.12GB

Original model description: --- basemodel: distilgpt2 libraryname: transformers license: apache-2.0 tags:

  • —generatedfromtrainer model-index:
  • —name: CFDistilGPT 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. -->

CFDistilGPT

This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 2.9080

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: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —num_epochs: 20

Training results

Training LossEpochStepValidation Loss
No log1.01403.2381
No log2.02803.1353
No log3.04203.0809
3.23464.05603.0432
3.23465.07003.0202
3.23466.08402.9960
3.23467.09802.9784
2.92168.011202.9666
2.92169.012602.9535
2.921610.014002.9435
2.785611.015402.9330
2.785612.016802.9253
2.785613.018202.9221
2.785614.019602.9184
2.697915.021002.9159
2.697916.022402.9133
2.697917.023802.9096
2.647918.025202.9102
2.647919.026602.9080
2.647920.028002.9080

Framework versions

  • —Transformers 4.46.1
  • —Pytorch 2.4.0+cu121
  • —Datasets 3.0.1
  • —Tokenizers 0.20.1