1{
2 "_class_name": "ModelStarOLMhead",
3 "activation_function": "gelu_new",
4 "architectures": [
5 "ModelStarOLMhead"
6 ],
7 "attn_pdrop": 0.1,
8 "bos_token_id": 0,
9 "dtype": "float32",
10 "embd_pdrop": 0.1,
11 "eos_token_id": 1,
12 "initializer_range": 0.02,
13 "layer_norm_epsilon": 5e-3,
14 "model_type": "staro",
15 "n_ctx": 512,
16 "n_embd": 1024,
17 "n_head": 16,
18 "n_inner": null,
19 "n_layer": 6,
20 "n_positions": 1024,
21 "pad_token_id": 1,
22 "reorder_and_upcast_attn": false,
23 "resid_pdrop": 0.1,
24 "scale_attn_by_inverse_layer_idx": false,
25 "scale_attn_weights": true,
26 "summary_activation": null,
27 "summary_first_dropout": 0.1,
28 "summary_proj_to_labels": true,
29 "summary_type": "cls_index",
30 "summary_use_proj": true,
31 "transformers_version": "4.57.3",
32 "use_cache": true,
33 "vocab_size": 60058,
34 "num_parameters": 192092160
35}
36ملف الإعدادات الكامل قديم من 2.65 سوبر
37{
38 "_class_name": "ModelStarOLMhead",
39 "activation_function": "gelu_new",
40 "architectures": [
41 "ModelStarOLMhead"
42 ],
43 "attn_pdrop": 0.1,
44 "bos_token_id": 0,
45 "dtype": "float32",
46 "embd_pdrop": 0.1,
47 "eos_token_id": 1,
48 "initializer_range": 0.02,
49 "layer_norm_epsilon": 1e-05,
50 "model_type": "staro",
51 "n_ctx": 512,
52 "n_embd": 1024,
53 "n_head": 16,
54 "n_inner": null,
55 "n_layer": 6,
56 "n_positions": 1024,
57 "pad_token_id": 1,
58 "reorder_and_upcast_attn": false,
59 "resid_pdrop": 0.1,
60 "scale_attn_by_inverse_layer_idx": false,
61 "scale_attn_weights": true,
62 "summary_activation": null,
63 "summary_first_dropout": 0.1,
64 "summary_proj_to_labels": true,
65 "summary_type": "cls_index",
66 "summary_use_proj": true,
67 "transformers_version": "4.57.3",
68 "use_cache": true,
69 "vocab_size": 56662
70}
71<details>
72
73عرض ملف الإعدادات الكامل (JSON) أقدم</summary>
74
75```json
76{
77 "_class_name": "ModelStarOLMhead",
78 "activation_function": "gelu_new",
79 "architectures": ["ModelStarOLMhead"],
80 "attn_pdrop": 0.1,
81 "bos_token_id": 0,
82 "dtype": "float32",
83 "embd_pdrop": 0.1,
84 "eos_token_id": 1,
85 "initializer_range": 0.02,
86 "layer_norm_epsilon": 1e-05,
87 "model_type": "staro",
88 "n_ctx": 512,
89 "n_embd": 1024,
90 "n_head": 16,
91 "n_inner": null,
92 "n_layer": 10,
93 "n_positions": 1024,
94 "pad_token_id": 1,
95 "resid_pdrop": 0.1,
96 "vocab_size": 40000
97}
98
99
100
101