Upload folder using huggingface_hub
Browse files- config.json +7 -7
- generation_config.json +3 -3
- model-00001-of-00002.safetensors +2 -2
- model-00002-of-00002.safetensors +2 -2
- model.safetensors.index.json +19 -19
- special_tokens_map.json +23 -17
- tokenizer.json +12 -44
- tokenizer_config.json +8 -28
config.json
CHANGED
@@ -1,16 +1,15 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "Zyphra/Zamba2-2.7B",
|
3 |
"add_bias_linear": false,
|
4 |
"architectures": [
|
5 |
"Zamba2ForCausalLM"
|
6 |
],
|
7 |
"attention_dropout": 0.0,
|
8 |
-
"bos_token_id":
|
9 |
"conv_dimension": 4,
|
10 |
-
"eos_token_id":
|
11 |
"expansion_factor": 2,
|
12 |
"ffn_hidden_size": 10240,
|
13 |
-
"ft_lora": false,
|
14 |
"gated_linear_unit": true,
|
15 |
"hidden_size": 2560,
|
16 |
"initializer_range": 0.02,
|
@@ -72,6 +71,7 @@
|
|
72 |
"m"
|
73 |
],
|
74 |
"lora_rank": 128,
|
|
|
75 |
"mamba_headdim": 64,
|
76 |
"max_position_embeddings": 4096,
|
77 |
"model_type": "zamba2",
|
@@ -81,18 +81,18 @@
|
|
81 |
"num_logits_to_keep": 1,
|
82 |
"num_mem_blocks": 2,
|
83 |
"num_query_groups": 32,
|
84 |
-
"pad_token_id":
|
85 |
"rms_norm_eps": 1e-05,
|
86 |
"rope_theta": 10000,
|
87 |
"se_shared_attention_lora": false,
|
88 |
"sliding_window": null,
|
89 |
"state_size": 64,
|
90 |
-
"torch_dtype": "
|
91 |
"transformers_version": "4.43.0.dev0",
|
92 |
"use_cache": false,
|
93 |
"use_mamba_kernels": true,
|
94 |
"use_mem_rope": false,
|
95 |
"use_shared_attention_lora": false,
|
96 |
"use_shared_block_lora": true,
|
97 |
-
"vocab_size":
|
98 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "Zyphra/Zamba2-2.7B-instruct",
|
3 |
"add_bias_linear": false,
|
4 |
"architectures": [
|
5 |
"Zamba2ForCausalLM"
|
6 |
],
|
7 |
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 1,
|
9 |
"conv_dimension": 4,
|
10 |
+
"eos_token_id": 2,
|
11 |
"expansion_factor": 2,
|
12 |
"ffn_hidden_size": 10240,
|
|
|
13 |
"gated_linear_unit": true,
|
14 |
"hidden_size": 2560,
|
15 |
"initializer_range": 0.02,
|
|
|
71 |
"m"
|
72 |
],
|
73 |
"lora_rank": 128,
|
74 |
+
"lora_rank_mamba": 128,
|
75 |
"mamba_headdim": 64,
|
76 |
"max_position_embeddings": 4096,
|
77 |
"model_type": "zamba2",
|
|
|
81 |
"num_logits_to_keep": 1,
|
82 |
"num_mem_blocks": 2,
|
83 |
"num_query_groups": 32,
|
84 |
+
"pad_token_id": 0,
|
85 |
"rms_norm_eps": 1e-05,
|
86 |
"rope_theta": 10000,
|
87 |
"se_shared_attention_lora": false,
|
88 |
"sliding_window": null,
|
89 |
"state_size": 64,
|
90 |
+
"torch_dtype": "float32",
|
91 |
"transformers_version": "4.43.0.dev0",
|
92 |
"use_cache": false,
|
93 |
"use_mamba_kernels": true,
|
94 |
"use_mem_rope": false,
|
95 |
"use_shared_attention_lora": false,
|
96 |
"use_shared_block_lora": true,
|
97 |
+
"vocab_size": 32000
|
98 |
}
|
generation_config.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"_from_model_config": true,
|
3 |
-
"bos_token_id":
|
4 |
-
"eos_token_id":
|
5 |
-
"pad_token_id":
|
6 |
"transformers_version": "4.43.0.dev0"
|
7 |
}
|
|
|
1 |
{
|
2 |
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"pad_token_id": 0,
|
6 |
"transformers_version": "4.43.0.dev0"
|
7 |
}
|
model-00001-of-00002.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02bfb07b081a97e16ba1222a980a4cb7bf5071981da538b0ed8247566773d870
|
3 |
+
size 4998338056
|
model-00002-of-00002.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:feb5b364c9438d060ab89931bd62c7065e32495d8cd2c5aff6bea5baf5d1a724
|
3 |
+
size 543078152
|
model.safetensors.index.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
"metadata": {
|
3 |
-
"total_size":
|
4 |
},
|
5 |
"weight_map": {
|
6 |
"model.blocks.0.feed_forward.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
@@ -453,7 +453,7 @@
|
|
453 |
"model.mamba_layers.47.mamba.in_proj.0.weight": "model-00001-of-00002.safetensors",
|
454 |
"model.mamba_layers.47.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
455 |
"model.mamba_layers.47.mamba.out_proj.weight": "model-00001-of-00002.safetensors",
|
456 |
-
"model.mamba_layers.48.input_layernorm.weight": "model-
|
457 |
"model.mamba_layers.48.mamba.A_log": "model-00001-of-00002.safetensors",
|
458 |
"model.mamba_layers.48.mamba.D": "model-00001-of-00002.safetensors",
|
459 |
"model.mamba_layers.48.mamba.conv1d.bias": "model-00001-of-00002.safetensors",
|
@@ -461,16 +461,16 @@
|
|
461 |
"model.mamba_layers.48.mamba.dt_bias": "model-00001-of-00002.safetensors",
|
462 |
"model.mamba_layers.48.mamba.in_proj.0.weight": "model-00001-of-00002.safetensors",
|
463 |
"model.mamba_layers.48.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
464 |
-
"model.mamba_layers.48.mamba.out_proj.weight": "model-
|
465 |
-
"model.mamba_layers.49.input_layernorm.weight": "model-
|
466 |
-
"model.mamba_layers.49.mamba.A_log": "model-
|
467 |
-
"model.mamba_layers.49.mamba.D": "model-
|
468 |
-
"model.mamba_layers.49.mamba.conv1d.bias": "model-
|
469 |
-
"model.mamba_layers.49.mamba.conv1d.weight": "model-
|
470 |
-
"model.mamba_layers.49.mamba.dt_bias": "model-
|
471 |
-
"model.mamba_layers.49.mamba.in_proj.0.weight": "model-
|
472 |
-
"model.mamba_layers.49.mamba.norm.weight": "model-
|
473 |
-
"model.mamba_layers.49.mamba.out_proj.weight": "model-
|
474 |
"model.mamba_layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
475 |
"model.mamba_layers.5.mamba.A_log": "model-00001-of-00002.safetensors",
|
476 |
"model.mamba_layers.5.mamba.D": "model-00001-of-00002.safetensors",
|
@@ -481,13 +481,13 @@
|
|
481 |
"model.mamba_layers.5.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
482 |
"model.mamba_layers.5.mamba.out_proj.weight": "model-00001-of-00002.safetensors",
|
483 |
"model.mamba_layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
484 |
-
"model.mamba_layers.50.mamba.A_log": "model-
|
485 |
-
"model.mamba_layers.50.mamba.D": "model-
|
486 |
-
"model.mamba_layers.50.mamba.conv1d.bias": "model-
|
487 |
-
"model.mamba_layers.50.mamba.conv1d.weight": "model-
|
488 |
-
"model.mamba_layers.50.mamba.dt_bias": "model-
|
489 |
-
"model.mamba_layers.50.mamba.in_proj.0.weight": "model-
|
490 |
-
"model.mamba_layers.50.mamba.norm.weight": "model-
|
491 |
"model.mamba_layers.50.mamba.out_proj.weight": "model-00002-of-00002.safetensors",
|
492 |
"model.mamba_layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
493 |
"model.mamba_layers.51.mamba.A_log": "model-00002-of-00002.safetensors",
|
|
|
1 |
{
|
2 |
"metadata": {
|
3 |
+
"total_size": 5541353280
|
4 |
},
|
5 |
"weight_map": {
|
6 |
"model.blocks.0.feed_forward.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
|
|
453 |
"model.mamba_layers.47.mamba.in_proj.0.weight": "model-00001-of-00002.safetensors",
|
454 |
"model.mamba_layers.47.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
455 |
"model.mamba_layers.47.mamba.out_proj.weight": "model-00001-of-00002.safetensors",
|
456 |
+
"model.mamba_layers.48.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
457 |
"model.mamba_layers.48.mamba.A_log": "model-00001-of-00002.safetensors",
|
458 |
"model.mamba_layers.48.mamba.D": "model-00001-of-00002.safetensors",
|
459 |
"model.mamba_layers.48.mamba.conv1d.bias": "model-00001-of-00002.safetensors",
|
|
|
461 |
"model.mamba_layers.48.mamba.dt_bias": "model-00001-of-00002.safetensors",
|
462 |
"model.mamba_layers.48.mamba.in_proj.0.weight": "model-00001-of-00002.safetensors",
|
463 |
"model.mamba_layers.48.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
464 |
+
"model.mamba_layers.48.mamba.out_proj.weight": "model-00002-of-00002.safetensors",
|
465 |
+
"model.mamba_layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
466 |
+
"model.mamba_layers.49.mamba.A_log": "model-00002-of-00002.safetensors",
|
467 |
+
"model.mamba_layers.49.mamba.D": "model-00002-of-00002.safetensors",
|
468 |
+
"model.mamba_layers.49.mamba.conv1d.bias": "model-00002-of-00002.safetensors",
|
469 |
+
"model.mamba_layers.49.mamba.conv1d.weight": "model-00002-of-00002.safetensors",
|
470 |
+
"model.mamba_layers.49.mamba.dt_bias": "model-00002-of-00002.safetensors",
|
471 |
+
"model.mamba_layers.49.mamba.in_proj.0.weight": "model-00002-of-00002.safetensors",
|
472 |
+
"model.mamba_layers.49.mamba.norm.weight": "model-00002-of-00002.safetensors",
|
473 |
+
"model.mamba_layers.49.mamba.out_proj.weight": "model-00002-of-00002.safetensors",
|
474 |
"model.mamba_layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
475 |
"model.mamba_layers.5.mamba.A_log": "model-00001-of-00002.safetensors",
|
476 |
"model.mamba_layers.5.mamba.D": "model-00001-of-00002.safetensors",
|
|
|
481 |
"model.mamba_layers.5.mamba.norm.weight": "model-00001-of-00002.safetensors",
|
482 |
"model.mamba_layers.5.mamba.out_proj.weight": "model-00001-of-00002.safetensors",
|
483 |
"model.mamba_layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
484 |
+
"model.mamba_layers.50.mamba.A_log": "model-00002-of-00002.safetensors",
|
485 |
+
"model.mamba_layers.50.mamba.D": "model-00002-of-00002.safetensors",
|
486 |
+
"model.mamba_layers.50.mamba.conv1d.bias": "model-00002-of-00002.safetensors",
|
487 |
+
"model.mamba_layers.50.mamba.conv1d.weight": "model-00002-of-00002.safetensors",
|
488 |
+
"model.mamba_layers.50.mamba.dt_bias": "model-00002-of-00002.safetensors",
|
489 |
+
"model.mamba_layers.50.mamba.in_proj.0.weight": "model-00002-of-00002.safetensors",
|
490 |
+
"model.mamba_layers.50.mamba.norm.weight": "model-00002-of-00002.safetensors",
|
491 |
"model.mamba_layers.50.mamba.out_proj.weight": "model-00002-of-00002.safetensors",
|
492 |
"model.mamba_layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
493 |
"model.mamba_layers.51.mamba.A_log": "model-00002-of-00002.safetensors",
|
special_tokens_map.json
CHANGED
@@ -1,23 +1,29 @@
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
3 |
-
|
4 |
-
|
5 |
-
"lstrip": false,
|
6 |
-
"normalized": false,
|
7 |
-
"rstrip": false,
|
8 |
-
"single_word": false
|
9 |
-
},
|
10 |
-
{
|
11 |
-
"content": "<|im_end|>",
|
12 |
-
"lstrip": false,
|
13 |
-
"normalized": false,
|
14 |
-
"rstrip": false,
|
15 |
-
"single_word": false
|
16 |
-
}
|
17 |
],
|
18 |
-
"bos_token":
|
19 |
-
|
20 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
"unk_token": {
|
22 |
"content": "<unk>",
|
23 |
"lstrip": false,
|
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
3 |
+
"<|im_start|>",
|
4 |
+
"<|im_end|>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<|im_start|>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<|im_end|>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "[PAD]",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
"unk_token": {
|
28 |
"content": "<unk>",
|
29 |
"lstrip": false,
|
tokenizer.json
CHANGED
@@ -1,42 +1,10 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
"direction": "Right",
|
5 |
-
"max_length": 4096,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
12 |
"id": 0,
|
13 |
-
"content": "<unk>",
|
14 |
-
"single_word": false,
|
15 |
-
"lstrip": false,
|
16 |
-
"rstrip": false,
|
17 |
-
"normalized": false,
|
18 |
-
"special": true
|
19 |
-
},
|
20 |
-
{
|
21 |
-
"id": 1,
|
22 |
-
"content": "<s>",
|
23 |
-
"single_word": false,
|
24 |
-
"lstrip": false,
|
25 |
-
"rstrip": false,
|
26 |
-
"normalized": false,
|
27 |
-
"special": true
|
28 |
-
},
|
29 |
-
{
|
30 |
-
"id": 2,
|
31 |
-
"content": "</s>",
|
32 |
-
"single_word": false,
|
33 |
-
"lstrip": false,
|
34 |
-
"rstrip": false,
|
35 |
-
"normalized": false,
|
36 |
-
"special": true
|
37 |
-
},
|
38 |
-
{
|
39 |
-
"id": 32000,
|
40 |
"content": "[PAD]",
|
41 |
"single_word": false,
|
42 |
"lstrip": false,
|
@@ -45,7 +13,7 @@
|
|
45 |
"special": true
|
46 |
},
|
47 |
{
|
48 |
-
"id":
|
49 |
"content": "<|im_start|>",
|
50 |
"single_word": false,
|
51 |
"lstrip": false,
|
@@ -54,7 +22,7 @@
|
|
54 |
"special": true
|
55 |
},
|
56 |
{
|
57 |
-
"id":
|
58 |
"content": "<|im_end|>",
|
59 |
"single_word": false,
|
60 |
"lstrip": false,
|
@@ -85,7 +53,7 @@
|
|
85 |
"single": [
|
86 |
{
|
87 |
"SpecialToken": {
|
88 |
-
"id": "
|
89 |
"type_id": 0
|
90 |
}
|
91 |
},
|
@@ -99,7 +67,7 @@
|
|
99 |
"pair": [
|
100 |
{
|
101 |
"SpecialToken": {
|
102 |
-
"id": "
|
103 |
"type_id": 0
|
104 |
}
|
105 |
},
|
@@ -111,7 +79,7 @@
|
|
111 |
},
|
112 |
{
|
113 |
"SpecialToken": {
|
114 |
-
"id": "
|
115 |
"type_id": 1
|
116 |
}
|
117 |
},
|
@@ -123,13 +91,13 @@
|
|
123 |
}
|
124 |
],
|
125 |
"special_tokens": {
|
126 |
-
"
|
127 |
-
"id": "
|
128 |
"ids": [
|
129 |
1
|
130 |
],
|
131 |
"tokens": [
|
132 |
-
"
|
133 |
]
|
134 |
}
|
135 |
}
|
@@ -168,9 +136,9 @@
|
|
168 |
"byte_fallback": true,
|
169 |
"ignore_merges": false,
|
170 |
"vocab": {
|
171 |
-
"
|
172 |
-
"
|
173 |
-
"
|
174 |
"<0x00>": 3,
|
175 |
"<0x01>": 4,
|
176 |
"<0x02>": 5,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
"content": "[PAD]",
|
9 |
"single_word": false,
|
10 |
"lstrip": false,
|
|
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
16 |
+
"id": 1,
|
17 |
"content": "<|im_start|>",
|
18 |
"single_word": false,
|
19 |
"lstrip": false,
|
|
|
22 |
"special": true
|
23 |
},
|
24 |
{
|
25 |
+
"id": 2,
|
26 |
"content": "<|im_end|>",
|
27 |
"single_word": false,
|
28 |
"lstrip": false,
|
|
|
53 |
"single": [
|
54 |
{
|
55 |
"SpecialToken": {
|
56 |
+
"id": "<|im_start|>",
|
57 |
"type_id": 0
|
58 |
}
|
59 |
},
|
|
|
67 |
"pair": [
|
68 |
{
|
69 |
"SpecialToken": {
|
70 |
+
"id": "<|im_start|>",
|
71 |
"type_id": 0
|
72 |
}
|
73 |
},
|
|
|
79 |
},
|
80 |
{
|
81 |
"SpecialToken": {
|
82 |
+
"id": "<|im_start|>",
|
83 |
"type_id": 1
|
84 |
}
|
85 |
},
|
|
|
91 |
}
|
92 |
],
|
93 |
"special_tokens": {
|
94 |
+
"<|im_start|>": {
|
95 |
+
"id": "<|im_start|>",
|
96 |
"ids": [
|
97 |
1
|
98 |
],
|
99 |
"tokens": [
|
100 |
+
"<|im_start|>"
|
101 |
]
|
102 |
}
|
103 |
}
|
|
|
136 |
"byte_fallback": true,
|
137 |
"ignore_merges": false,
|
138 |
"vocab": {
|
139 |
+
"[PAD]": 0,
|
140 |
+
"<|im_start|>": 1,
|
141 |
+
"<|im_end|>": 2,
|
142 |
"<0x00>": 3,
|
143 |
"<0x01>": 4,
|
144 |
"<0x02>": 5,
|
tokenizer_config.json
CHANGED
@@ -4,30 +4,6 @@
|
|
4 |
"add_prefix_space": null,
|
5 |
"added_tokens_decoder": {
|
6 |
"0": {
|
7 |
-
"content": "<unk>",
|
8 |
-
"lstrip": false,
|
9 |
-
"normalized": false,
|
10 |
-
"rstrip": false,
|
11 |
-
"single_word": false,
|
12 |
-
"special": true
|
13 |
-
},
|
14 |
-
"1": {
|
15 |
-
"content": "<s>",
|
16 |
-
"lstrip": false,
|
17 |
-
"normalized": false,
|
18 |
-
"rstrip": false,
|
19 |
-
"single_word": false,
|
20 |
-
"special": true
|
21 |
-
},
|
22 |
-
"2": {
|
23 |
-
"content": "</s>",
|
24 |
-
"lstrip": false,
|
25 |
-
"normalized": false,
|
26 |
-
"rstrip": false,
|
27 |
-
"single_word": false,
|
28 |
-
"special": true
|
29 |
-
},
|
30 |
-
"32000": {
|
31 |
"content": "[PAD]",
|
32 |
"lstrip": false,
|
33 |
"normalized": false,
|
@@ -35,7 +11,7 @@
|
|
35 |
"single_word": false,
|
36 |
"special": true
|
37 |
},
|
38 |
-
"
|
39 |
"content": "<|im_start|>",
|
40 |
"lstrip": false,
|
41 |
"normalized": false,
|
@@ -43,7 +19,7 @@
|
|
43 |
"single_word": false,
|
44 |
"special": true
|
45 |
},
|
46 |
-
"
|
47 |
"content": "<|im_end|>",
|
48 |
"lstrip": false,
|
49 |
"normalized": false,
|
@@ -56,16 +32,20 @@
|
|
56 |
"<|im_start|>",
|
57 |
"<|im_end|>"
|
58 |
],
|
59 |
-
"bos_token": "
|
60 |
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
61 |
"clean_up_tokenization_spaces": false,
|
62 |
-
"eos_token": "
|
63 |
"legacy": true,
|
|
|
64 |
"model_max_length": 4096,
|
65 |
"pad_token": "[PAD]",
|
66 |
"sp_model_kwargs": {},
|
67 |
"spaces_between_special_tokens": false,
|
|
|
68 |
"tokenizer_class": "LlamaTokenizer",
|
|
|
|
|
69 |
"unk_token": "<unk>",
|
70 |
"use_default_system_prompt": false
|
71 |
}
|
|
|
4 |
"add_prefix_space": null,
|
5 |
"added_tokens_decoder": {
|
6 |
"0": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
"content": "[PAD]",
|
8 |
"lstrip": false,
|
9 |
"normalized": false,
|
|
|
11 |
"single_word": false,
|
12 |
"special": true
|
13 |
},
|
14 |
+
"1": {
|
15 |
"content": "<|im_start|>",
|
16 |
"lstrip": false,
|
17 |
"normalized": false,
|
|
|
19 |
"single_word": false,
|
20 |
"special": true
|
21 |
},
|
22 |
+
"2": {
|
23 |
"content": "<|im_end|>",
|
24 |
"lstrip": false,
|
25 |
"normalized": false,
|
|
|
32 |
"<|im_start|>",
|
33 |
"<|im_end|>"
|
34 |
],
|
35 |
+
"bos_token": "<|im_start|>",
|
36 |
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
37 |
"clean_up_tokenization_spaces": false,
|
38 |
+
"eos_token": "<|im_end|>",
|
39 |
"legacy": true,
|
40 |
+
"max_length": 4096,
|
41 |
"model_max_length": 4096,
|
42 |
"pad_token": "[PAD]",
|
43 |
"sp_model_kwargs": {},
|
44 |
"spaces_between_special_tokens": false,
|
45 |
+
"stride": 0,
|
46 |
"tokenizer_class": "LlamaTokenizer",
|
47 |
+
"truncation_side": "right",
|
48 |
+
"truncation_strategy": "longest_first",
|
49 |
"unk_token": "<unk>",
|
50 |
"use_default_system_prompt": false
|
51 |
}
|