text_encoder/config.json

{
  "architectures": [
    "Qwen3ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "dtype": "bfloat16",
  "eos_token_id": 151645,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 9728,
  "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"
  ],
  "max_position_embeddings": 40960,
  "max_window_layers": 36,
  "model_type": "qwen3",
  "num_attention_heads": 32,
  "num_hidden_layers": 27,
  "num_key_value_heads": 8,
  "pad_token_id": null,
  "quantization_config": {
    "include_input_output_embeddings": false,
    "modules_to_not_convert": [
      "lm_head"
    ],
    "quant_method": "torchao",
    "quant_type": {
      "default": {
        "_data": {
          "step": null,
          "use_dynamic_per_tensor_scale": true,
          "use_triton_kernel": true
        },
        "_type": "NVFP4DynamicActivationNVFP4WeightConfig",
        "_version": 1
      }
    },
    "untie_embedding_weights": false
  },
  "rms_norm_eps": 1e-06,
  "rope_parameters": {
    "rope_theta": 1000000,
    "rope_type": "default"
  },
  "sliding_window": null,
  "tie_word_embeddings": true,
  "transformers_version": "5.14.0.dev0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}