Skip to page content
SAVRN
Search Plan a project

Open-weight model · Text classification

Toxic-Not-Toxic

by Saber Bounehas saberbounehas/Toxic-Not-Toxic

Toxic-Not-Toxic is an open-weight model for text classification from Saber Bounehas, released under Apache License 2.0. It has 82M parameters and a 514-token context. At 16-bit it needs about 0.2 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.

This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 8 - evalbatchsize: 8 …

Parameters82M
Context514
Weights328.5 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—

Runs On

What it takes to serve Toxic-Not-Toxic (82M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 0.2 GB 0.2 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.1 GB 0.1 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.0 GB 0.0 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.

Toxic-Not-Toxic on every accelerator the SAVRN Index prices, at every precision

Model Card

By Saber Bounehas, published under apache-2.0, revision 649bdb970e3d.

This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 10 - Transformers 5.19.0 - Pytorch 2.14.1 - Datasets 5.1.0 - Tokenizers 0.23.2

Read Saber Bounehas's full model card

This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0169 - Accuracy: 0.91 - Precision: 0.9301 - Recall: 0.8867 - F1: 0.9078

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: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1
0.0000 1.0 88 1.1921 0.8733 0.8077 0.98 0.8855
0.0000 2.0 176 0.9924 0.8933 0.9538 0.8267 0.8857
0.0000 3.0 264 0.8227 0.9267 0.9156 0.94 0.9276
0.0000 4.0 352 0.8310 0.9267 0.9156 0.94 0.9276
0.0000 5.0 440 0.8379 0.9267 0.9156 0.94 0.9276
0.0000 6.0 528 0.8446 0.9267 0.9156 0.94 0.9276
0.0000 7.0 616 1.0147 0.91 0.9301 0.8867 0.9078
0.0000 8.0 704 1.0174 0.91 0.9301 0.8867 0.9078
0.0000 9.0 792 1.0165 0.91 0.9301 0.8867 0.9078
0.0000 10.0 880 1.0169 0.91 0.9301 0.8867 0.9078

Framework versions

  • Transformers 5.19.0
  • Pytorch 2.14.1
  • Datasets 5.1.0
  • Tokenizers 0.23.2

Configuration

Architecture
RobertaForSequenceClassification
Context length (tokens)
514
Layers
6
Hidden size
768
Feed-forward size
3,072
Attention heads
12
Vocabulary size
50,265
Model type
roberta

Identity and Version

Repository
saberbounehas/Toxic-Not-Toxic
Publisher
Saber Bounehas
Task
Text classification
Modality
Text
Library
transformers
Parameters
82M parameters
Languages
Not stated by the source
Revision
649bdb970e3d2f196c0a1d4ccd133277d2f0700e
First published
2026-10-11
Last updated
2026-10-11

Files and Weights

7 files, 332.1 MB in total. The weights are 2 files totalling 328.5 MB in bin, safetensors.

Weights2 files · 328.5 MB
Configuration1 file · 806 B
Tokenizer2 files · 3.6 MB
Documentation1 file · 2.4 KB
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights328.5 MB bf7bb0975f24
training_args.binWeights5.2 KB 5e2585295197
config.jsonConfiguration806 B —
README.mdDocumentation2.4 KB —
.gitattributesRepository1.5 KB —
tokenizer.jsonTokenizer3.6 MB —
tokenizer_config.jsonTokenizer388 B —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
328.5 MB
Download from Saber Bounehas

Released by Saber Bounehas through its official repository on Hugging Face. Read the license.

Built From

Memory Requirements

PrecisionWeights in memory
As published328.5 MB
16-bit0.2 GB
8-bit0.1 GB
4-bit0.0 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About Toxic-Not-Toxic

How much GPU memory does Toxic-Not-Toxic need?

About 0.2 GB at 16-bit and 0 GB at 4-bit: the weights (82M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run Toxic-Not-Toxic on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use Toxic-Not-Toxic commercially?

Yes. Toxic-Not-Toxic is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is Toxic-Not-Toxic's context length?

514 tokens, from the maximum position embeddings in its published configuration.

Similar Models

This model is a fine-tuned version of distilroberta-base on the financialphrasebank dataset. It achieves the following results on the evaluation set: This model is a distilled version of the RoBERTa-base model. It follows the same training procedure as DistilBERT. The code for the distillation process can be found here. The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 125M parameters for RoBERTa-base). On average DistilRoBERTa is twice as fast as Roberta-base. Polar sentiment dataset of sentences from financial news. The dataset consists of 4840 sentences from English language financial news categorised by sentiment. The dataset is divided by…

Open weights apache-2.0 82M parameters 514 tokens transformers
View model

Model · Text classification

gpt-news-model

Mohit Raj Nayak

This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 16 - evalbatchsize: 16 - lrschedulertype: linear - numepochs: 3 - Transformers 5.16.1 - Pytorch 2.11.0+cu128 - Datasets 4.8.5 - Tokenizers 0.23.1

Open weights apache-2.0 82M parameters transformers
View model

Model · Text classification

gpt-news-model

Aman Ankit

This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 16 - evalbatchsize: 16 - lrschedulertype: linear - numepochs: 3 - Transformers 5.17.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2

Open weights apache-2.0 82M parameters transformers
View model

Model · Text classification

kobert

Kong

This model is a fine-tuned version of skt/kobert-base-v1 on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 16 - evalbatchsize: 16 - lrschedulertype: linear - numepochs: 5 - Transformers 5.17.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2

Open weights 92M parameters 512 tokens transformers
View model

Model · Text classification

Llama-Prompt-Guard-2-22M-onnx

AthithyaSakthivel8

This repository provides a ONNX converted and quantized version of meta-llama/Llama-Prompt-Guard-2-22M - Meta LLaMA – Foundation model powering the classifier - Hugging Face Transformers – Model and tokenizer loading - ONNX – Model export and runtime format - ONNX Runtime – Efficient inference backend We use jackhhao/jailbreak-classification for the evaluation (train+test) You can find the full source code, CLI tools, and evaluation scripts in the official GitHub repository.

Open weights llama4 71M parameters 512 tokens
View model

Model · Text classification

scb-banking77-distilbert-test-3

CK Tsai

This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Open weights 67M parameters 512 tokens transformers
View model