Modelpublic

marrodion/minilm-l12-v2-toxic-ft

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes75downloads

Model Card

SentenceTransformer based on marrodion/minilm-l12-v2-simple

This is a sentence-transformers model finetuned from marrodion/minilm-l12-v2-simple. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

—Model Type: Sentence Transformer
—Base model: marrodion/minilm-l12-v2-simple
—Maximum Sequence Length: 128 tokens
—Output Dimensionality: 384 tokens
—Similarity Function: Cosine Similarity

Model Sources

—Documentation: Sentence Transformers Documentation
—Repository: Sentence Transformers on GitHub
—Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

bash

pip install -U sentence-transformers

Then you can load this model and run inference.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("marrodion/minilm-l12-v2-toxic-ft")
# Run inference
sentences = [
    'eradicate the gays',
    'Anyone who supports ISIS terrorism globally should be eradicated.',
    'This movie was filled with garbage Zionist propaganda.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!--

Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details> -->

<!--

Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details> -->

<!--

Out-of-Scope Use

List how the model may foreseeably be misused and address what users ought not to do with the model. -->

Evaluation

Metrics

Semantic Similarity

—Dataset: hatespeech-sampled-dev
—Evaluated with <code>EmbeddingSimilarityEvaluator</code>

Metric	Value
pearson_cosine	0.5825
spearman_cosine	0.4527
pearson_manhattan	0.5684
spearman_manhattan	0.4501
pearson_euclidean	0.57
spearman_euclidean	0.4527
pearson_dot	0.5825
spearman_dot	0.4527
pearson_max	0.5825
spearman_max	0.4527

<!--

Bias, Risks and Limitations

What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. -->

<!--

Recommendations

What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. -->

Training Details

Training Dataset

Unnamed Dataset

—Size: 33,827 training samples
—Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
—Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | score | |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------| | type | string | string | float | | details | <ul><li>min: 4 tokens</li><li>mean: 35.5 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 34.35 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.82</li><li>max: 1.0</li></ul> |
—Samples: | sentence1 | sentence2 | score | |:----------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------|:--------------------------------| | <code>Don't waste good pussy on a trash ass nigga</code> | <code>Stfu obviously trans people aren't retarded</code> | <code>0.7777777777777778</code> | | <code>The more I read about dumb violent things do, like beating up two women bc they wouldn't kiss for them.. the gayer I become #Pride2019</code> | <code>Nuke them all</code> | <code>0.5555555555555556</code> | | <code>Forget her cunt and ass and fuck her face till she pukes!</code> | <code>@QTAnon1 Hope she fucks her all the way back to Somalia..</code> | <code>1.0</code> |
—Loss: <code>CosineSimilarityLoss</code> with these parameters:

json

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }

Evaluation Dataset

Unnamed Dataset

—Size: 1,781 evaluation samples
—Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
—Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | score | |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------| | type | string | string | float | | details | <ul><li>min: 4 tokens</li><li>mean: 35.77 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 36.67 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 0.11</li><li>mean: 0.83</li><li>max: 1.0</li></ul> |
—Samples: | sentence1 | sentence2 | score | |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------| | <code>So for Trump and his inner circle to be crying over law and order, they sure know how to follow the system huh? Yet nothing about his ways of following the law were legal. So, how is his asylum rules were ever legal if children were sick and forced to live in unconditioned areas is American or right?</code> | <code>We need every illegal alien gang member in prison. Then deported.</code> | <code>1.0</code> | | <code>hey girl are you my dick cause im about to beat the shit out of you</code> | <code>Reading all these hot takes from Americans, they all just seem like utter dogs. A human mind should not be this deranged, yet this is the norm in America</code> | <code>0.8888888888888888</code> | | <code>Its amazing what a failure women are in politics. Its almost like its not in their nature to be powerful. No matter how much they try, poor dears....</code> | <code>Tbh that cut makes him look like neonazi. He looks like he's about to lecture me about how white people are the most discriminates against racially after he called me the n word with that hard r.</code> | <code>0.8888888888888888</code> |
—Loss: <code>CosineSimilarityLoss</code> with these parameters:

json

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }

Training Hyperparameters

Non-Default Hyperparameters

—eval_strategy: steps
—per_device_train_batch_size: 32
—per_device_eval_batch_size: 64
—num_train_epochs: 1
—warmup_ratio: 0.1
—load_best_model_at_end: True

All Hyperparameters

<details><summary>Click to expand</summary>

—overwrite_output_dir: False
—do_predict: False
—eval_strategy: steps
—prediction_loss_only: True
—per_device_train_batch_size: 32
—per_device_eval_batch_size: 64
—per_gpu_train_batch_size: None
—per_gpu_eval_batch_size: None
—gradient_accumulation_steps: 1
—eval_accumulation_steps: None
—learning_rate: 5e-05
—weight_decay: 0.0
—adam_beta1: 0.9
—adam_beta2: 0.999
—adam_epsilon: 1e-08
—max_grad_norm: 1.0
—num_train_epochs: 1
—max_steps: -1
—lr_scheduler_type: linear
—lr_scheduler_kwargs: {}
—warmup_ratio: 0.1
—warmup_steps: 0
—log_level: passive
—log_level_replica: warning
—log_on_each_node: True
—logging_nan_inf_filter: True
—save_safetensors: True
—save_on_each_node: False
—save_only_model: False
—restore_callback_states_from_checkpoint: False
—no_cuda: False
—use_cpu: False
—use_mps_device: False
—seed: 42
—data_seed: None
—jit_mode_eval: False
—use_ipex: False
—bf16: False
—fp16: False
—fp16_opt_level: O1
—half_precision_backend: auto
—bf16_full_eval: False
—fp16_full_eval: False
—tf32: None
—local_rank: 0
—ddp_backend: None
—tpu_num_cores: None
—tpu_metrics_debug: False
—debug: []
—dataloader_drop_last: False
—dataloader_num_workers: 0
—dataloader_prefetch_factor: None
—past_index: -1
—disable_tqdm: False
—remove_unused_columns: True
—label_names: None
—load_best_model_at_end: True
—ignore_data_skip: False
—fsdp: []
—fsdp_min_num_params: 0
—fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
—fsdp_transformer_layer_cls_to_wrap: None
—accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
—deepspeed: None
—label_smoothing_factor: 0.0
—optim: adamw_torch
—optim_args: None
—adafactor: False
—group_by_length: False
—length_column_name: length
—ddp_find_unused_parameters: None
—ddp_bucket_cap_mb: None
—ddp_broadcast_buffers: False
—dataloader_pin_memory: True
—dataloader_persistent_workers: False
—skip_memory_metrics: True
—use_legacy_prediction_loop: False
—push_to_hub: False
—resume_from_checkpoint: None
—hub_model_id: None
—hub_strategy: every_save
—hub_private_repo: False
—hub_always_push: False
—gradient_checkpointing: False
—gradient_checkpointing_kwargs: None
—include_inputs_for_metrics: False
—eval_do_concat_batches: True
—fp16_backend: auto
—push_to_hub_model_id: None
—push_to_hub_organization: None
—mp_parameters:
—auto_find_batch_size: False
—full_determinism: False
—torchdynamo: None
—ray_scope: last
—ddp_timeout: 1800
—torch_compile: False
—torch_compile_backend: None
—torch_compile_mode: None
—dispatch_batches: None
—split_batches: None
—include_tokens_per_second: False
—include_num_input_tokens_seen: False
—neftune_noise_alpha: None
—optim_target_modules: None
—batch_eval_metrics: False
—batch_sampler: batch_sampler
—multi_dataset_batch_sampler: proportional

</details>

Training Logs

Epoch	Step	Training Loss	loss	hatespeech-sampled-dev_spearman_cosine
0.2836	300	0.0503	0.0139	0.4258
0.5671	600	0.0143	0.0135	0.4418
0.8507	900	0.0134	0.0131	0.4527

—The bold row denotes the saved checkpoint.

Framework Versions

—Python: 3.10.14
—Sentence Transformers: 3.0.0
—Transformers: 4.41.1
—PyTorch: 2.3.0
—Accelerate: 0.30.1
—Datasets: 2.19.1
—Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

bibtex

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

<!--

Glossary

Clearly define terms in order to be accessible across audiences. -->

<!--

Model Card Authors

Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. -->

<!--

Model Card Contact

Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->