Upload 3 files
Browse filesadd config and tokenizer of VRS-HQ
- config.json +53 -0
- generation_config.json +16 -0
- tokenizer.model +3 -0
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/18515601223/chat_univi",
|
| 3 |
+
"aarchitectures": "ChatUniViLlamaForCausalLM",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"VISAForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"config": {
|
| 9 |
+
"freeze": false,
|
| 10 |
+
"mm_tune": true,
|
| 11 |
+
"spatial_cluster_rate0": 64,
|
| 12 |
+
"spatial_cluster_rate1": 32,
|
| 13 |
+
"spatial_cluster_rate2": 16,
|
| 14 |
+
"temporal_cluster_rate": 0.0625,
|
| 15 |
+
"use_cluster": true,
|
| 16 |
+
"vision_tune": false
|
| 17 |
+
},
|
| 18 |
+
"eos_token_id": 2,
|
| 19 |
+
"freeze_mm_mlp_adapter": true,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 4096,
|
| 22 |
+
"image_aspect_ratio": "square",
|
| 23 |
+
"image_grid_pinpoints": null,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 11008,
|
| 26 |
+
"max_position_embeddings": 4096,
|
| 27 |
+
"mm_hidden_size": 1024,
|
| 28 |
+
"mm_use_box_start_end": false,
|
| 29 |
+
"mm_use_im_patch_token": false,
|
| 30 |
+
"mm_use_im_start_end": true,
|
| 31 |
+
"mm_vision_select_feature": "patch",
|
| 32 |
+
"mm_vision_select_layer": -2,
|
| 33 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14",
|
| 34 |
+
"model_type": "ChatUniVi",
|
| 35 |
+
"num_attention_heads": 32,
|
| 36 |
+
"num_hidden_layers": 32,
|
| 37 |
+
"num_key_value_heads": 32,
|
| 38 |
+
"out_dim": 256,
|
| 39 |
+
"pad_token_id": 0,
|
| 40 |
+
"pretrain_mm_mlp_adapter": null,
|
| 41 |
+
"pretraining_tp": 1,
|
| 42 |
+
"rms_norm_eps": 1e-05,
|
| 43 |
+
"rope_scaling": null,
|
| 44 |
+
"tie_word_embeddings": false,
|
| 45 |
+
"torch_dtype": "float32",
|
| 46 |
+
"train_mask_decoder": true,
|
| 47 |
+
"transformers_version": "4.31.0",
|
| 48 |
+
"tune_mm_mlp_adapter": false,
|
| 49 |
+
"use_cache": false,
|
| 50 |
+
"use_mm_proj": true,
|
| 51 |
+
"vision_tower": "openai/clip-vit-large-patch14",
|
| 52 |
+
"vocab_size": 32002
|
| 53 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha": 0.1,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"max_length": 4096,
|
| 6 |
+
"out_dim": 256,
|
| 7 |
+
"pad_token_id": 0,
|
| 8 |
+
"seg_token_idx": 32000,
|
| 9 |
+
"seg_token_num": 1,
|
| 10 |
+
"temperature": 0.9,
|
| 11 |
+
"top_p": 0.6,
|
| 12 |
+
"track_token_idx": 32001,
|
| 13 |
+
"train_mask_decoder": true,
|
| 14 |
+
"transformers_version": "4.31.0",
|
| 15 |
+
"vision_tower": "openai/clip-vit-large-patch14"
|
| 16 |
+
}
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|