RichardErkhov/Cambridge-KAIST2_-_SmolLM-14m-Dolma-v0.1-Base-tied-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SmolLM-14m-Dolma-v0.1-Base-tied - GGUF

• —Model creator: https://huggingface.co/Cambridge-KAIST2/
• —Original model: https://huggingface.co/Cambridge-KAIST2/SmolLM-14m-Dolma-v0.1-Base-tied/

Original model description: --- libraryname: transformers basemodel: JW17/SmolLM-14m-v0.1 tags:

• —generatedfromtrainer model-index:
• —name: SmolLM-14m-Dolma-v0.1-Base-tied 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. -->

SmolLM-14m-Dolma-v0.1-Base-tied

This model is a fine-tuned version of JW17/SmolLM-14m-v0.1 on an unknown dataset.

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: 0.003
• —trainbatchsize: 1024
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 2
• —totaltrainbatch_size: 2048
• —totalevalbatch_size: 16
• —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: warmupstabledecay
• —lrschedulerwarmup_steps: 1000
• —num_epochs: 1

Training results

Framework versions

• —Transformers 4.47.1
• —Pytorch 2.5.1
• —Datasets 3.2.0
• —Tokenizers 0.21.0