LanguageBind's picture
Upload 11 files
532e5ac verified
{
"_name_or_path": "./checkpoints/llavaphi-3b-other-384",
"architectures": [
"MoELLaVAPhiForCausalLM"
],
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "microsoft/phi-2--configuration_phi.PhiConfig",
"AutoModelForCausalLM": "microsoft/phi-2--modeling_phi.PhiForCausalLM"
},
"bos_token_id": null,
"embd_pdrop": 0.0,
"eos_token_id": null,
"freeze_mm_mlp_adapter": false,
"hidden_act": "gelu_new",
"hidden_size": 2560,
"image_aspect_ratio": "pad",
"image_projector_type": "mlp2x_gelu",
"initializer_range": 0.02,
"intermediate_size": 10240,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 2048,
"mm_hidden_size": 1152,
"mm_image_tower": "google/siglip-so400m-patch14-384",
"mm_projector_lr": null,
"mm_use_im_patch_token": false,
"mm_use_im_start_end": false,
"mm_video_tower": null,
"mm_vision_select_feature": "patch",
"mm_vision_select_layer": -2,
"model_type": "moe_llava_phi",
"moe": {
"capacity_factor": 1.5,
"ep_size": 1,
"eval_capacity_factor": 2.0,
"min_capacity": 0,
"moe_enable": true,
"moe_layers_idx": [
0,
2,
4,
6,
8,
10,
12,
14,
16,
18,
20,
22,
24,
26,
28,
30
],
"moe_mode": "sparse",
"num_experts": [
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4,
4
],
"router_aux_loss_coef": 0.01,
"top_k_experts": 2,
"train_modules": [
"fc1",
"fc2",
"wg"
],
"use_residual": false
},
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 32,
"pad_token_id": 50295,
"partial_rotary_factor": 0.4,
"qk_layernorm": false,
"resid_pdrop": 0.1,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"tokenizer_padding_side": "right",
"torch_dtype": "bfloat16",
"transformers_version": "4.37.0",
"tune_mm_mlp_adapter": false,
"use_cache": true,
"use_mm_proj": true,
"video_global_proj": false,
"video_projector_type": "linear",
"video_spatial_proj": false,
"video_temproal_proj": false,
"vocab_size": 51200
}