Update config.json
Browse files- config.json +53 -2
config.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"_name_or_path": "LeroyDyer/Mixtral_AI_CyberBrain_3_0",
|
3 |
"architectures": [
|
4 |
-
"MistralForCausalLM"
|
5 |
],
|
6 |
"attention_dropout": 0.0,
|
7 |
"bos_token_id": 1,
|
@@ -11,6 +11,46 @@
|
|
11 |
"initializer_range": 0.02,
|
12 |
"intermediate_size": 14336,
|
13 |
"max_position_embeddings": 32768,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
"max_thoughts": 10,
|
15 |
"merged_lm_and_talk_heads": false,
|
16 |
"merged_lm_and_think_heads": true,
|
@@ -21,10 +61,21 @@
|
|
21 |
"num_key_value_heads": 8,
|
22 |
"rms_norm_eps": 1e-05,
|
23 |
"rope_theta": 10000.0,
|
24 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
"tie_word_embeddings": false,
|
26 |
"torch_dtype": "float16",
|
27 |
"transformers_version": "4.40.0.dev0",
|
|
|
|
|
|
|
|
|
28 |
"use_cache": true,
|
29 |
"use_complex_talk_head": true,
|
30 |
"use_complex_think_head": false,
|
|
|
1 |
{
|
2 |
"_name_or_path": "LeroyDyer/Mixtral_AI_CyberBrain_3_0",
|
3 |
"architectures": [
|
4 |
+
"MistralForCausalLM","LlavaMistralForCausalLM"
|
5 |
],
|
6 |
"attention_dropout": 0.0,
|
7 |
"bos_token_id": 1,
|
|
|
11 |
"initializer_range": 0.02,
|
12 |
"intermediate_size": 14336,
|
13 |
"max_position_embeddings": 32768,
|
14 |
+
"freeze_mm_mlp_adapter": false,
|
15 |
+
"freeze_mm_vision_resampler": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "anyres",
|
19 |
+
"image_crop_resolution": 224,
|
20 |
+
"image_grid_pinpoints": [
|
21 |
+
[
|
22 |
+
336,
|
23 |
+
672
|
24 |
+
],
|
25 |
+
[
|
26 |
+
672,
|
27 |
+
336
|
28 |
+
],
|
29 |
+
[
|
30 |
+
672,
|
31 |
+
672
|
32 |
+
],
|
33 |
+
[
|
34 |
+
1008,
|
35 |
+
336
|
36 |
+
],
|
37 |
+
[
|
38 |
+
336,
|
39 |
+
1008
|
40 |
+
]
|
41 |
+
],
|
42 |
+
"image_split_resolution": 224,
|
43 |
+
"mm_hidden_size": 1024,
|
44 |
+
"mm_patch_merge_type": "spatial_unpad",
|
45 |
+
"mm_projector_lr": null,
|
46 |
+
"mm_projector_type": "mlp2x_gelu",
|
47 |
+
"mm_resampler_type": null,
|
48 |
+
"mm_use_im_patch_token": false,
|
49 |
+
"mm_use_im_start_end": false,
|
50 |
+
"mm_vision_select_feature": "patch",
|
51 |
+
"mm_vision_select_layer": -2,
|
52 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
53 |
+
"mm_vision_tower_lr": 2e-06,
|
54 |
"max_thoughts": 10,
|
55 |
"merged_lm_and_talk_heads": false,
|
56 |
"merged_lm_and_think_heads": true,
|
|
|
61 |
"num_key_value_heads": 8,
|
62 |
"rms_norm_eps": 1e-05,
|
63 |
"rope_theta": 10000.0,
|
64 |
+
"rope_scaling": {
|
65 |
+
"factor": 16.0,
|
66 |
+
"finetuned": true,
|
67 |
+
"original_max_position_embeddings": 8192,
|
68 |
+
"type": "yarn"
|
69 |
+
},
|
70 |
+
"rope_theta": 10000.0,
|
71 |
+
"sliding_window": 131072,
|
72 |
"tie_word_embeddings": false,
|
73 |
"torch_dtype": "float16",
|
74 |
"transformers_version": "4.40.0.dev0",
|
75 |
+
"tune_mm_mlp_adapter": false,
|
76 |
+
"tune_mm_vision_resampler": false,
|
77 |
+
"unfreeze_mm_vision_tower": true,
|
78 |
+
"use_mm_proj": true,
|
79 |
"use_cache": true,
|
80 |
"use_complex_talk_head": true,
|
81 |
"use_complex_think_head": false,
|