Update config.json
Browse files- config.json +2 -86
config.json
CHANGED
@@ -1,108 +1,24 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "heegyu/mistralai__Mistral-7B-v0.1-mistral-7b-v0.1-lima",
|
3 |
"architectures": [
|
4 |
"MistralForCausalLM"
|
5 |
],
|
6 |
-
"attention_bias": false,
|
7 |
-
"attention_dropout": 0.0,
|
8 |
-
"attention_partition_spec": [
|
9 |
-
[
|
10 |
-
"dp",
|
11 |
-
"fsdp"
|
12 |
-
],
|
13 |
-
"sp",
|
14 |
-
"tp",
|
15 |
-
null
|
16 |
-
],
|
17 |
-
"attn_mechanism": "normal",
|
18 |
-
"axis_dims": [
|
19 |
-
1,
|
20 |
-
-1,
|
21 |
-
1,
|
22 |
-
1
|
23 |
-
],
|
24 |
-
"axis_names": [
|
25 |
-
"dp",
|
26 |
-
"fsdp",
|
27 |
-
"tp",
|
28 |
-
"sp"
|
29 |
-
],
|
30 |
-
"backend": null,
|
31 |
-
"bias_partition_spec": [
|
32 |
-
[
|
33 |
-
"dp",
|
34 |
-
"fsdp"
|
35 |
-
],
|
36 |
-
null,
|
37 |
-
null,
|
38 |
-
null
|
39 |
-
],
|
40 |
-
"bits": null,
|
41 |
-
"block_b": 1,
|
42 |
-
"block_k": 128,
|
43 |
-
"block_k_dkv": 128,
|
44 |
-
"block_k_dq": 128,
|
45 |
-
"block_k_major": 128,
|
46 |
-
"block_k_major_dkv": 128,
|
47 |
-
"block_k_major_dq": 128,
|
48 |
-
"block_q": 128,
|
49 |
-
"block_q_dkv": 128,
|
50 |
-
"block_q_dq": 128,
|
51 |
-
"block_q_major_dkv": 128,
|
52 |
"bos_token_id": 1,
|
53 |
-
"easy_method": "train",
|
54 |
"eos_token_id": 2,
|
55 |
"hidden_act": "silu",
|
56 |
"hidden_size": 4096,
|
57 |
"initializer_range": 0.02,
|
58 |
"intermediate_size": 14336,
|
59 |
-
"key_partition_spec": [
|
60 |
-
[
|
61 |
-
"dp",
|
62 |
-
"fsdp"
|
63 |
-
],
|
64 |
-
"sp",
|
65 |
-
"tp",
|
66 |
-
null
|
67 |
-
],
|
68 |
"max_position_embeddings": 32768,
|
69 |
"model_type": "mistral",
|
70 |
"num_attention_heads": 32,
|
71 |
"num_hidden_layers": 32,
|
72 |
"num_key_value_heads": 8,
|
73 |
-
"number_rep_kv": 1,
|
74 |
-
"query_partition_spec": [
|
75 |
-
[
|
76 |
-
"dp",
|
77 |
-
"fsdp"
|
78 |
-
],
|
79 |
-
"sp",
|
80 |
-
"tp",
|
81 |
-
null
|
82 |
-
],
|
83 |
"rms_norm_eps": 1e-05,
|
84 |
-
"rope_scaling": null,
|
85 |
"rope_theta": 10000.0,
|
86 |
-
"scan_attention_layers": false,
|
87 |
-
"scan_mlp_chunk_size": 1024,
|
88 |
-
"scan_ring_attention": true,
|
89 |
"sliding_window": 4096,
|
90 |
"tie_word_embeddings": false,
|
91 |
"torch_dtype": "bfloat16",
|
92 |
-
"transformers_version": "4.
|
93 |
"use_cache": true,
|
94 |
-
"use_pjit_attention_force": false,
|
95 |
-
"use_scan_mlp": false,
|
96 |
-
"use_shard_map": false,
|
97 |
-
"use_sharded_kv_caching": true,
|
98 |
-
"value_partition_spec": [
|
99 |
-
[
|
100 |
-
"dp",
|
101 |
-
"fsdp"
|
102 |
-
],
|
103 |
-
"sp",
|
104 |
-
"tp",
|
105 |
-
null
|
106 |
-
],
|
107 |
"vocab_size": 32000
|
108 |
-
}
|
|
|
1 |
{
|
|
|
2 |
"architectures": [
|
3 |
"MistralForCausalLM"
|
4 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"bos_token_id": 1,
|
|
|
6 |
"eos_token_id": 2,
|
7 |
"hidden_act": "silu",
|
8 |
"hidden_size": 4096,
|
9 |
"initializer_range": 0.02,
|
10 |
"intermediate_size": 14336,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
"max_position_embeddings": 32768,
|
12 |
"model_type": "mistral",
|
13 |
"num_attention_heads": 32,
|
14 |
"num_hidden_layers": 32,
|
15 |
"num_key_value_heads": 8,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
"rms_norm_eps": 1e-05,
|
|
|
17 |
"rope_theta": 10000.0,
|
|
|
|
|
|
|
18 |
"sliding_window": 4096,
|
19 |
"tie_word_embeddings": false,
|
20 |
"torch_dtype": "bfloat16",
|
21 |
+
"transformers_version": "4.34.0.dev0",
|
22 |
"use_cache": true,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
"vocab_size": 32000
|
24 |
+
}
|