| { |
| "architectures": [ |
| "GraniteSpeechForConditionalGeneration" |
| ], |
| "audio_token_index": 49159, |
| "downsample_rate": 5, |
| "encoder_config": { |
| "context_size": 200, |
| "conv_expansion_factor": 2, |
| "conv_kernel_size": 15, |
| "dim_head": 128, |
| "dropout": 0.1, |
| "feedforward_mult": 4, |
| "hidden_dim": 1024, |
| "input_dim": 160, |
| "max_pos_emb": 512, |
| "model_type": "granite_speech_encoder", |
| "num_heads": 8, |
| "num_layers": 16, |
| "output_dim": 256, |
| "torch_dtype": "float16" |
| }, |
| "has_lora_adapter": true, |
| "initializer_range": 0.02, |
| "model_type": "granite_speech", |
| "projector_config": { |
| "attention_probs_dropout_prob": 0.1, |
| "cross_attention_frequency": 1, |
| "encoder_hidden_size": 1024, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 1024, |
| "initializer_range": 0.02, |
| "intermediate_size": 4096, |
| "layer_norm_eps": 1e-12, |
| "max_position_embeddings": 2048, |
| "model_type": "blip_2_qformer", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 2, |
| "position_embedding_type": "absolute", |
| "torch_dtype": "float16", |
| "use_qformer_text_input": false, |
| "vocab_size": 30522 |
| }, |
| "quantization_config": { |
| "_load_in_4bit": true, |
| "_load_in_8bit": false, |
| "bnb_4bit_compute_dtype": "bfloat16", |
| "bnb_4bit_quant_storage": "uint8", |
| "bnb_4bit_quant_type": "nf4", |
| "bnb_4bit_use_double_quant": true, |
| "llm_int8_enable_fp32_cpu_offload": false, |
| "llm_int8_has_fp16_weight": false, |
| "llm_int8_skip_modules": null, |
| "llm_int8_threshold": 6.0, |
| "load_in_4bit": true, |
| "load_in_8bit": false, |
| "quant_method": "bitsandbytes" |
| }, |
| "text_config": { |
| "_name_or_path": "/proj/speech/data/granite-3.3-8b-instruct/r250409a", |
| "architectures": [ |
| "GraniteForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attention_multiplier": 0.0078125, |
| "bos_token_id": 0, |
| "embedding_multiplier": 12.0, |
| "eos_token_id": 0, |
| "hidden_act": "silu", |
| "hidden_size": 4096, |
| "initializer_range": 0.02, |
| "intermediate_size": 12800, |
| "logits_scaling": 16.0, |
| "max_position_embeddings": 131072, |
| "mlp_bias": false, |
| "model_type": "granite", |
| "num_attention_heads": 32, |
| "num_hidden_layers": 40, |
| "num_key_value_heads": 8, |
| "pad_token_id": 0, |
| "residual_multiplier": 0.22, |
| "rms_norm_eps": 1e-05, |
| "rope_scaling": null, |
| "rope_theta": 10000000.0, |
| "tie_word_embeddings": true, |
| "torch_dtype": "float16", |
| "use_cache": true, |
| "vocab_size": 49160 |
| }, |
| "torch_dtype": "float16", |
| "transformers_version": "4.53.1", |
| "window_size": 15 |
| } |
|
|
