mirror of
https://github.com/mcloh/decision-models-benchmark.git
synced 2026-11-06 06:14:11 +00:00
328ccd1c2e
- harness/evaluate.py: partição congelada com SHA-256, amostra com seed, lote 1, truncamento comum, contagem de tokens, latência (sem aquecimento), pico de RAM e GPU; teste bloqueado até o congelamento da configuração. - analysis/metrics.py: acurácia, macro F1, Brier, NLL, ECE, acurácia seletiva. - Diagnósticos por resposta (metadata.diagnostics); Laya informa opções colapsadas pelo orçamento de cabeçalho; head_max_len via DMB_LAYA_HEAD_MAX_LEN. - infra/ds/collect.py e gpu_queue --env para pilotos e rodadas. - .gitignore: de models/ só metadados (cards, configs, licenças, índices); pesos, tokenizadores grandes, runtimes, fontes clonados e caches ignorados. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
80 lines
1.9 KiB
JSON
80 lines
1.9 KiB
JSON
{
|
|
"architectures": [
|
|
"ModernBertForMaskedLM"
|
|
],
|
|
"attention_bias": false,
|
|
"attention_dropout": 0.0,
|
|
"bos_token_id": 2,
|
|
"classifier_activation": "gelu",
|
|
"classifier_bias": false,
|
|
"classifier_dropout": 0.0,
|
|
"classifier_pooling": "mean",
|
|
"cls_token_id": 1,
|
|
"decoder_bias": true,
|
|
"deterministic_flash_attn": false,
|
|
"dtype": "float32",
|
|
"embedding_dropout": 0.0,
|
|
"eos_token_id": 1,
|
|
"global_attn_every_n_layers": 3,
|
|
"gradient_checkpointing": false,
|
|
"hidden_activation": "gelu",
|
|
"hidden_size": 768,
|
|
"initializer_cutoff_factor": 2.0,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 1152,
|
|
"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,
|
|
"mask_token_id": 4,
|
|
"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": 0,
|
|
"position_embedding_type": "sans_pos",
|
|
"repad_logits_with_grad": false,
|
|
"rope_parameters": {
|
|
"full_attention": {
|
|
"rope_theta": 160000,
|
|
"rope_type": "default"
|
|
},
|
|
"sliding_attention": {
|
|
"rope_theta": 160000,
|
|
"rope_type": "default"
|
|
}
|
|
},
|
|
"sep_token_id": 1,
|
|
"sparse_pred_ignore_index": -100,
|
|
"sparse_prediction": false,
|
|
"tie_word_embeddings": true,
|
|
"transformers_version": "5.0.0",
|
|
"vocab_size": 256000
|
|
}
|