Safetensors
modernbert
Androcles-2 / config.json
steee's picture
Upload 5 files
06c4dbc verified
Raw History Blame Contribute Delete
3 kB
{
"architectures": [
"ModernBertForSequenceClassification"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": null,
"classifier_activation": "gelu",
"classifier_bias": false,
"classifier_dropout": 0.0,
"classifier_pooling": "mean",
"cls_token_id": 50281,
"decoder_bias": true,
"deterministic_flash_attn": false,
"dtype": "float32",
"embedding_dropout": 0.0,
"eos_token_id": null,
"global_attn_every_n_layers": 3,
"gradient_checkpointing": false,
"hidden_activation": "gelu",
"hidden_size": 768,
"id2label": {
"0": "Brave",
"1": "Browser",
"2": "Choice",
"3": "Coding",
"4": "Data Analysis",
"5": "Diagrams",
"6": "Fact Checking",
"7": "Finance",
"8": "Image Generation",
"9": "Math / Calculations",
"10": "Multilingualism",
"11": "News",
"12": "Recommendation",
"13": "Sports",
"14": "Structured Writing",
"15": "Summarisation",
"16": "Thinking",
"17": "Time-critical",
"18": "Translation",
"19": "Travel Planning",
"20": "Weather"
},
"initializer_cutoff_factor": 2.0,
"initializer_range": 0.02,
"intermediate_size": 1152,
"label2id": {
"Brave": 0,
"Browser": 1,
"Choice": 2,
"Coding": 3,
"Data Analysis": 4,
"Diagrams": 5,
"Fact Checking": 6,
"Finance": 7,
"Image Generation": 8,
"Math / Calculations": 9,
"Multilingualism": 10,
"News": 11,
"Recommendation": 12,
"Sports": 13,
"Structured Writing": 14,
"Summarisation": 15,
"Thinking": 16,
"Time-critical": 17,
"Translation": 18,
"Travel Planning": 19,
"Weather": 20
},
"layer_norm_eps": 1e-05,
"layer_types": [
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"local_attention": 128,
"max_position_embeddings": 8192,
"mlp_bias": false,
"mlp_dropout": 0.0,
"model_type": "modernbert",
"norm_bias": false,
"norm_eps": 1e-05,
"num_attention_heads": 12,
"num_hidden_layers": 22,
"pad_token_id": 50283,
"position_embedding_type": "absolute",
"problem_type": "multi_label_classification",
"rope_parameters": {
"full_attention": {
"rope_theta": 160000.0,
"rope_type": "default"
},
"sliding_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"sep_token_id": 50282,
"sparse_pred_ignore_index": -100,
"sparse_prediction": false,
"tie_word_embeddings": true,
"transformers_version": "5.7.0",
"use_cache": false,
"vocab_size": 50368
}