Hugging Face's logo

tencent
/
HiLS-Attention-7B

Text Generation
Transformers
Safetensors
English
olmo_hils
long-context
sparse-attention
efficient-attention
pretraining
olmo3
hils-attention
Model card Files
xet
 Community
imhuim982 commited on
Commit
c64f89d
·
verified ·
1 Parent(s): d10bf9a

Upload baseline checkpoints: nsa_rope_345M

Browse files
Files changed (7) hide show
  1. baseline_ckpts/nsa_rope_345M/config.json +54 -0
  2. baseline_ckpts/nsa_rope_345M/merges.txt +0 -0
  3. baseline_ckpts/nsa_rope_345M/model.safetensors +3 -0
  4. baseline_ckpts/nsa_rope_345M/special_tokens_map.json +30 -0
  5. baseline_ckpts/nsa_rope_345M/tokenizer.json +0 -0
  6. baseline_ckpts/nsa_rope_345M/tokenizer_config.json +190 -0
  7. baseline_ckpts/nsa_rope_345M/vocab.json +0 -0
baseline_ckpts/nsa_rope_345M/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 100257,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 100257,
10
+ "full_attn_interleave": 1,
11
+ "head_dim": 64,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 4096,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention"
33
+ ],
34
+ "max_position_embeddings": 131072,
35
+ "max_window_layers": 16,
36
+ "model_type": "qwen3",
37
+ "nsa_block_counts": 32,
38
+ "nsa_block_size": 64,
39
+ "nsa_num_key_value_heads": 2,
40
+ "nsa_use_rope": true,
41
+ "nsa_window_size": 512,
42
+ "num_attention_heads": 16,
43
+ "num_hidden_layers": 16,
44
+ "num_key_value_heads": 2,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_scaling": null,
47
+ "rope_theta": 10000.0,
48
+ "sliding_window": 512,
49
+ "tie_word_embeddings": true,
50
+ "transformers_version": "4.57.3",
51
+ "use_cache": true,
52
+ "use_sliding_window": true,
53
+ "vocab_size": 100278
54
+ }
baseline_ckpts/nsa_rope_345M/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
baseline_ckpts/nsa_rope_345M/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4ad58fe8a992aeb7c94e5db3e5c5ce7fb69fca78309e832cb1c2622ef05ec4c
3
+ size 890548288
baseline_ckpts/nsa_rope_345M/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|endoftext|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|pad|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
baseline_ckpts/nsa_rope_345M/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
baseline_ckpts/nsa_rope_345M/tokenizer_config.json ADDED
@@ -0,0 +1,190 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "100256": {
5
+ "content": "<|system|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": false
11
+ },
12
+ "100257": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "100258": {
21
+ "content": "<|fim_prefix|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "100259": {
29
+ "content": "<|fim_middle|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "100260": {
37
+ "content": "<|fim_suffix|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "100261": {
45
+ "content": "|||PHONE_NUMBER|||",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": false
51
+ },
52
+ "100262": {
53
+ "content": "|||EMAIL_ADDRESS|||",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": false
59
+ },
60
+ "100263": {
61
+ "content": "|||IP_ADDRESS|||",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": false
67
+ },
68
+ "100264": {
69
+ "content": "<|im_start|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "100265": {
77
+ "content": "<|im_end|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "100266": {
85
+ "content": "<|user|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": false
91
+ },
92
+ "100267": {
93
+ "content": "<|assistant|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": false
99
+ },
100
+ "100268": {
101
+ "content": "<think>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": false
107
+ },
108
+ "100269": {
109
+ "content": "</think>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": false
115
+ },
116
+ "100270": {
117
+ "content": "<functions>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "100271": {
125
+ "content": "</functions>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "100272": {
133
+ "content": "<function_calls>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "100273": {
141
+ "content": "</function_calls>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "100274": {
149
+ "content": "<answer>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "100275": {
157
+ "content": "</answer>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "100276": {
165
+ "content": "<|endofprompt|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": true
171
+ },
172
+ "100277": {
173
+ "content": "<|pad|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": true
179
+ }
180
+ },
181
+ "bos_token": "<|endoftext|>",
182
+ "clean_up_tokenization_spaces": false,
183
+ "eos_token": "<|endoftext|>",
184
+ "extra_special_tokens": {},
185
+ "model_max_length": 8192,
186
+ "pad_token": "<|pad|>",
187
+ "padding_side": "right",
188
+ "tokenizer_class": "GPT2Tokenizer",
189
+ "unk_token": "<|endoftext|>"
190
+ }
baseline_ckpts/nsa_rope_345M/vocab.json ADDED
The diff for this file is too large to render. See raw diff