RichardErkhov/Trelis_-_99-base-checkpoints-v10-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

99-base-checkpoints-v10 - GGUF

• —Model creator: https://huggingface.co/Trelis/
• —Original model: https://huggingface.co/Trelis/99-base-checkpoints-v10/

Original model description: --- libraryname: transformers basemodel: Trelis/SmolLM-135M-layer-pruned-90M-raw tags:

• —trl
• —sft
• —generatedfromtrainer model-index:
• —name: 99-v10-base 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. -->

99-v10-base

This model is a fine-tuned version of Trelis/SmolLM-135M-layer-pruned-90M-raw on the None dataset. It achieves the following results on the evaluation set:

• —Loss: 0.9658

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: 0.001
• —trainbatchsize: 16
• —evalbatchsize: 16
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 8
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 512
• —totalevalbatch_size: 128
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —lrschedulerwarmup_ratio: 0.01
• —lrschedulerwarmup_steps: 19
• —training_steps: 1951

Training results

Framework versions

• —Transformers 4.44.2
• —Pytorch 2.1.1+cu121
• —Datasets 3.0.0
• —Tokenizers 0.19.1