Rahul Thapa
added files
bcb8d25
{
"_name_or_path": "/data/rahulthapa/multi-modality/checkpoints/dragonfly_v1_vit_L_meanpool_with_med/dragonfly_v1_vit_L_meanpool-stage2-ablation-model_for_full_training",
"architectures": [
"DragonflyForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 128000,
"eos_token_id": 128001,
"hidden_act": "silu",
"hidden_size": 4096,
"image_encoder": "openai/clip-vit-large-patch14-336",
"image_size": 300,
"initializer_range": 0.02,
"intermediate_size": 14336,
"max_position_embeddings": 8192,
"model_type": "dragonfly",
"num_attention_heads": 32,
"num_channels": 3,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"patch_size": 30,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 500000.0,
"text_config": {
"bos_token_id": 128000,
"eos_token_id": 128001,
"intermediate_size": 14336,
"max_position_embeddings": 8192,
"model_type": "llama",
"num_key_value_heads": 8,
"rms_norm_eps": 1e-05,
"rope_theta": 500000.0,
"vocab_size": 128256
},
"text_pretrained_model_name_or_path": "meta-llama/Meta-Llama-3.1-8B-Instruct",
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.44.2",
"use_cache": true,
"vocab_size": 128256
}