toxicity-scorer-smollm2-135m-freeze

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-135M on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3011
  • F1: 0.8565
  • Accuracy: 0.8807
  • Precision: 0.8558
  • Recall: 0.8807

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: 3e-05
  • train_batch_size: 36
  • eval_batch_size: 36
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 288
  • total_eval_batch_size: 288
  • 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 1.2430 0.5996 0.5195 0.7569 0.5195
0.3157 1.2762 5000 0.3011 0.8562 0.8806 0.8556 0.8806
0.309 2.5523 10000 0.3011 0.8565 0.8807 0.8558 0.8807

Framework versions

  • Transformers 4.46.3
  • Pytorch 2.5.1
  • Datasets 3.1.0
  • Tokenizers 0.20.3
Downloads last month
32
Safetensors
Model size
135M params
Tensor type
BF16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for tcapelle/toxicity-scorer-smollm2-135m-freeze

Finetuned
(163)
this model