README.md

metadata

library_name: transformers
license: apache-2.0
base_model: HuggingFaceTB/SmolLM2-360M
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
  - precision
  - recall
model-index:
  - name: toxicity-scorer-smollm2-360m-freeze
    results: []

toxicity-scorer-smollm2-360m-freeze

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-360M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2347
• F1: 0.9013
• Accuracy: 0.9033
• Precision: 0.9006
• Recall: 0.9033

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy	Precision	Recall
No log	0	0	0.8130	0.5487	0.5049	0.6645	0.5049
0.2957	0.2340	5000	0.2896	0.8803	0.8841	0.8795	0.8841
0.2451	0.4680	10000	0.2443	0.8976	0.8995	0.8968	0.8995
0.2349	0.7020	15000	0.2383	0.8994	0.9020	0.8989	0.9020
0.2277	0.9360	20000	0.2363	0.9006	0.9027	0.8999	0.9027
0.2414	1.1700	25000	0.2352	0.9013	0.9035	0.9007	0.9035
0.2361	1.4040	30000	0.2349	0.9013	0.9035	0.9007	0.9035
0.2312	1.6380	35000	0.2348	0.9013	0.9033	0.9007	0.9033
0.2207	1.8720	40000	0.2348	0.9014	0.9035	0.9007	0.9035
0.2645	2.1060	45000	0.2347	0.9012	0.9033	0.9005	0.9033
0.2369	2.3399	50000	0.2347	0.9012	0.9033	0.9005	0.9033
0.2329	2.5739	55000	0.2347	0.9013	0.9034	0.9006	0.9034
0.2253	2.8079	60000	0.2347	0.9013	0.9033	0.9006	0.9033

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3