ZaandaTeika's picture
Convert model to bfloat16 and fix total_parameters metadata
ce0ad6f verified
{
"architectures": [
"Qwen2ForTokenClassification"
],
"attention_dropout": 0.0,
"dtype": "bfloat16",
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 3584,
"id2label": {
"0": "error",
"1": "correct"
},
"initializer_range": 0.02,
"intermediate_size": 18944,
"label2id": {
"correct": 1,
"error": 0
},
"layer_types": [
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention"
],
"max_position_embeddings": 4096,
"max_window_layers": 28,
"model_type": "qwen2",
"num_attention_heads": 28,
"num_hidden_layers": 28,
"num_key_value_heads": 4,
"pad_token_id": 151643,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000.0,
"sliding_window": null,
"tie_word_embeddings": false,
"transformers_version": "4.56.2",
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 152064
}