Hugging Face's logo

yanolja
/
YanoljaNEXT-Rosetta-27B-2511

Translation
Transformers
Safetensors
gemma3_text
text-generation
text-generation-inference
Model card Files
xet
Community
seungduk commited on
Commit
125fe7c
·
verified ·
1 Parent(s): a62fc22

Add files using upload-large-folder tool

Browse files
Files changed (24) hide show
  1. .gitattributes +1 -0
  2. README.md +253 -0
  3. added_tokens.json +3 -0
  4. chat_template.jinja +22 -0
  5. config.json +101 -0
  6. generation_config.json +13 -0
  7. model-00001-of-00012.safetensors +3 -0
  8. model-00002-of-00012.safetensors +3 -0
  9. model-00003-of-00012.safetensors +3 -0
  10. model-00004-of-00012.safetensors +3 -0
  11. model-00005-of-00012.safetensors +3 -0
  12. model-00006-of-00012.safetensors +3 -0
  13. model-00007-of-00012.safetensors +3 -0
  14. model-00008-of-00012.safetensors +3 -0
  15. model-00009-of-00012.safetensors +3 -0
  16. model-00010-of-00012.safetensors +3 -0
  17. model-00011-of-00012.safetensors +3 -0
  18. model-00012-of-00012.safetensors +3 -0
  19. model.safetensors.index.json +817 -0
  20. special_tokens_map.json +30 -0
  21. tokenizer.json +3 -0
  22. tokenizer.model +3 -0
  23. tokenizer_config.json +0 -0
  24. wmt24pp_12b.md +39 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - translation
5
+ license: gemma
6
+ language:
7
+ - ar
8
+ - bg
9
+ - zh
10
+ - cs
11
+ - da
12
+ - nl
13
+ - en
14
+ - fi
15
+ - fr
16
+ - de
17
+ - el
18
+ - gu
19
+ - he
20
+ - hi
21
+ - hu
22
+ - id
23
+ - it
24
+ - ja
25
+ - ko
26
+ - fa
27
+ - pl
28
+ - pt
29
+ - ro
30
+ - ru
31
+ - sk
32
+ - es
33
+ - sv
34
+ - tl
35
+ - th
36
+ - tr
37
+ - uk
38
+ - vi
39
+ ---
40
+
41
+ # YanoljaNEXT-Rosetta-27B-2511
42
+
43
+ <p style="text-align: center; margin: 0 auto 64px">
44
+ <img src="next_rosetta.png" style="width: 1096px">
45
+ </p>
46
+
47
+ This model is a fine-tuned version of [`google/gemma-3-27b-pt`](https://huggingface.co/google/gemma-3-27b-pt). As it is intended solely for text generation, we have extracted and utilized only the `Gemma3ForCausalLM` component from the original architecture.
48
+
49
+ Unlike our previous EEVE models, this model does not feature an expanded tokenizer.
50
+
51
+ - **Model Name:** `yanolja/YanoljaNEXT-Rosetta-27B-2511`
52
+ - **Base Model:** `google/gemma-3-27b-pt`
53
+
54
+ ## Model Description
55
+
56
+ This model is a 27-billion parameter, decoder-only language model built on the Gemma3 architecture and fine-tuned by Yanolja NEXT. It is specifically designed to translate structured data (JSON, YAML, XML formats) while preserving the original data structure.
57
+
58
+ The model was trained on a multilingual dataset covering the following languages equally:
59
+ - Arabic
60
+ - Bulgarian
61
+ - Chinese
62
+ - Czech
63
+ - Danish
64
+ - Dutch
65
+ - English
66
+ - Finnish
67
+ - French
68
+ - German
69
+ - Greek
70
+ - Gujarati
71
+ - Hebrew
72
+ - Hindi
73
+ - Hungarian
74
+ - Indonesian
75
+ - Italian
76
+ - Japanese
77
+ - Korean
78
+ - Persian
79
+ - Polish
80
+ - Portuguese
81
+ - Romanian
82
+ - Russian
83
+ - Slovak
84
+ - Spanish
85
+ - Swedish
86
+ - Tagalog
87
+ - Thai
88
+ - Turkish
89
+ - Ukrainian
90
+ - Vietnamese
91
+
92
+ While optimized for these languages, it may also perform effectively on other languages supported by the base Gemma3 model.
93
+
94
+ ## How to use
95
+
96
+ You can use this model with the `transformers` library as follows:
97
+
98
+ ```python
99
+ import json
100
+ import torch
101
+ from transformers import AutoTokenizer, AutoModelForCausalLM
102
+
103
+ model_id = "yanolja/YanoljaNEXT-Rosetta-27B-2511"
104
+ model = AutoModelForCausalLM.from_pretrained(
105
+ model_id,
106
+ device_map="auto",
107
+ )
108
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
109
+
110
+ target_language = "Korean"
111
+ context = {
112
+ "context": "Simple introduction about a tech company.",
113
+ "tone": "Informative and helpful",
114
+ "glossary": {
115
+ "Yanolja NEXT": "야놀자넥스트",
116
+ "travel industry": "여행 산업",
117
+ }
118
+ }
119
+
120
+ system = [f"Translate the user's text to {target_language}."]
121
+ for key, value in context.items():
122
+ key_pascal = key.capitalize()
123
+ if isinstance(value, dict):
124
+ system.append(f"{key_pascal}:")
125
+ for f, t in value.items():
126
+ system.append(f"- {f} -> {t}")
127
+ else:
128
+ system.append(f"{key_pascal}: {value}")
129
+
130
+ system.append("Provide the final translation immediately without any other text.")
131
+
132
+ source = {
133
+ "company_name": "Yanolja NEXT",
134
+ "description": "Yanolja NEXT is a company that provides cutting-edge "
135
+ "technology for the global travel industry.",
136
+ }
137
+
138
+ messages = [
139
+ {"role": "system", "content": "\n".join(system)},
140
+ {"role": "user", "content": json.dumps(source, ensure_ascii=False)},
141
+ ]
142
+
143
+ prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
144
+ inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
145
+ input_length = inputs["input_ids"].shape[1]
146
+
147
+ with torch.inference_mode():
148
+ outputs = model.generate(
149
+ **inputs,
150
+ max_new_tokens=64,
151
+ )
152
+
153
+ generated_tokens = outputs[0][input_length:]
154
+ translation = tokenizer.decode(generated_tokens, skip_special_tokens=True)
155
+
156
+ print(json.dumps(json.loads(translation), indent=2, ensure_ascii=False))
157
+ ```
158
+
159
+ The model outputs the final translation in the same structured format as the input (JSON, YAML, XML) when appropriate, or plain text for simple translations.
160
+
161
+ ## Training Procedure
162
+
163
+ ### Training Data
164
+ The translation datasets were synthesized using fineweb corpora.
165
+ - [FineWeb Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu)
166
+ - [FineWeb2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2)
167
+
168
+ The model was fine-tuned on synthetic multilingual translation data to optimize performance across the supported language pairs.
169
+
170
+ ## Performance
171
+
172
+ ### Translation Quality Benchmarks
173
+
174
+ The following CHrF++ scores (WMT24++) demonstrate the model's competitive performance compared to other state-of-the-art translation models on English to Korean translation:
175
+
176
+ | Model | CHrF++ Score (WMT24++) |
177
+ |------------------------------------|--------------|
178
+ | **yanolja/YanoljaNEXT-Rosetta-27B-2511** | **37.21** |
179
+ | **yanolja/YanoljaNEXT-Rosetta-4B-2511** | **35.64** |
180
+ | google/gemini-2.5-flash-lite | 35.23 |
181
+ | yanolja/YanoljaNEXT-Rosetta-20B | 33.87 |
182
+ | google/gemini-2.0-flash-001 | 33.81 |
183
+ | openai/gpt-oss-120b | 31.51 |
184
+ | openai/gpt-4.1-nano | 31.15 |
185
+ | Qwen/Qwen3-235B-A22B-Instruct-2507-FP8 | 31.02 |
186
+ | openai/gpt-oss-20b | 30.56 |
187
+ | google/gemma-3-27b-it | 30.05 |
188
+
189
+ YanoljaNEXT-Rosetta-27B-2511 achieves strong translation quality while maintaining efficient inference for its parameter size.
190
+ Scores for the other language pairs can be found in the [WMT24++ Evaluation Results](wmt24pp_12b.md).
191
+
192
+ ## Intended Uses & Limitations
193
+
194
+ This model is intended for translating structured data (JSON, YAML, XML formats) while preserving the original structure. It is particularly well-suited for tasks such as localizing product catalogs, translating hotel reviews, or handling any other structured content that requires accurate translation.
195
+
196
+ ### Limitations
197
+ The model is primarily optimized for processing structured data (JSON, YAML, XML).
198
+ Its performance on unstructured text or other data formats may vary.
199
+ In some cases, the model may produce invalid structured outputs, repetitive output, or inaccurate translations.
200
+
201
+ ### License
202
+ This model is released under the Gemma license, inherited from its base architecture. Please consult the official [Gemma license terms](https://ai.google.dev/gemma/terms) for detailed usage guidelines.
203
+
204
+ ## Acknowledgments
205
+ This work was supported by the Korea Creative Content Agency (KOCCA) grant, funded by the Ministry of Culture, Sports and Tourism (MCST) in 2025 (Project Name: _Cultivating Masters and Doctoral Experts to Lead Digital-Tech Tourism_, Project Number: RS-2024-00442006, Contribution Rate: 100%).
206
+
207
+ ## Citation
208
+
209
+ If you use this model, please consider citing:
210
+
211
+ ```
212
+ @misc{yanolja2025yanoljanextrosetta27b,
213
+ author = {Yanolja NEXT Co., Ltd.},
214
+ title = {YanoljaNEXT-Rosetta-27B-2511},
215
+ year = {2025},
216
+ publisher = {Hugging Face},
217
+ journal = {Hugging Face repository},
218
+ howpublished = {\\url{https://huggingface.co/yanolja/YanoljaNEXT-Rosetta-27B-2511}}
219
+ }
220
+ ```
221
+
222
+ ## References
223
+
224
+ This work utilizes several models and datasets. We would like to acknowledge the original authors for their valuable contributions to the field.
225
+
226
+ ```
227
+ @misc{gemma3,
228
+ author = {Google},
229
+ title = {Gemma 3},
230
+ year = {2024},
231
+ publisher = {Google DeepMind},
232
+ howpublished = {\\url{https://deepmind.google/models/gemma/gemma-3/}}
233
+ }
234
+
235
+ @misc{penedo2025fineweb2pipelinescale,
236
+ title = {FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language},
237
+ author = {Guilherme Penedo and Hynek Kydlíček and Vinko Sabolčec and Bettina Messmer and Negar Foroutan and Amir Hossein Kargaran and Colin Raffel and Martin Jaggi and Leandro Von Werra and Thomas Wolf},
238
+ year = {2025},
239
+ eprint = {2506.20920},
240
+ archivePrefix = {arXiv},
241
+ primaryClass = {cs.CL},
242
+ url = {https://arxiv.org/abs/2506.20920},
243
+ }
244
+
245
+ @misc{lozhkov2024fineweb-edu,
246
+ author = {Lozhkov, Anton and Ben Allal, Loubna and von Werra, Leandro and Wolf, Thomas},
247
+ title = {FineWeb-Edu: the Finest Collection of Educational Content},
248
+ year = 2024,
249
+ url = {https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu},
250
+ doi = {10.57967/hf/2497},
251
+ publisher={Hugging Face}
252
+ }
253
+ ```
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
chat_template.jinja ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set system_msg = messages | selectattr('role', 'eq', 'system') | list | first | default(none) -%}
2
+ {%- set last_user = messages | selectattr('role', 'eq', 'user') | list | last | default(none) -%}
3
+ {%- set last_assistant = messages | selectattr('role', 'eq', 'assistant') | list | last | default(none) -%}
4
+
5
+ {{- bos_token -}}
6
+
7
+ {%- if system_msg -%}
8
+ <start_of_turn>instruction{{ '\n' }}
9
+ {{- system_msg['content'] | trim -}}<end_of_turn>{{ '\n' }}
10
+ {%- endif -%}
11
+
12
+ {%- if last_user -%}
13
+ <start_of_turn>source{{ '\n' }}
14
+ {{- last_user['content'] | trim -}}<end_of_turn>{{ '\n' }}
15
+ {%- endif -%}
16
+
17
+ {%- if add_generation_prompt -%}
18
+ <start_of_turn>translation{{ '\n' }}
19
+ {%- elif last_assistant -%}
20
+ <start_of_turn>translation{{ '\n' }}
21
+ {{- last_assistant['content'] | trim -}}<end_of_turn>{{ '\n' }}
22
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 106,
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 128,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 5376,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 21504,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "full_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "sliding_attention",
55
+ "full_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "sliding_attention",
61
+ "full_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "sliding_attention",
67
+ "full_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "sliding_attention",
73
+ "full_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "sliding_attention",
79
+ "full_attention",
80
+ "sliding_attention",
81
+ "sliding_attention"
82
+ ],
83
+ "max_position_embeddings": 131072,
84
+ "model_type": "gemma3_text",
85
+ "num_attention_heads": 32,
86
+ "num_hidden_layers": 62,
87
+ "num_key_value_heads": 16,
88
+ "pad_token_id": 0,
89
+ "query_pre_attn_scalar": 168,
90
+ "rms_norm_eps": 1e-06,
91
+ "rope_local_base_freq": 10000.0,
92
+ "rope_scaling": {
93
+ "factor": 8.0,
94
+ "rope_type": "linear"
95
+ },
96
+ "rope_theta": 1000000.0,
97
+ "sliding_window": 1024,
98
+ "transformers_version": "4.56.1",
99
+ "use_cache": false,
100
+ "vocab_size": 262208
101
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.56.1"
13
+ }
model-00001-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b011b89d1391a4faab1833e9778cee86bf78b5e3bd9f6e1fd3521a65f7971056
3
+ size 4834190712
model-00002-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e8b2e14bc65e3cff8b9f31b9944a0f99f95f3077599ca9232efc7c9967a728e
3
+ size 4954791768
model-00003-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bbf93a15e112105076d9513c13531b1e8c5a24878e93287ede48905ad42f267
3
+ size 4954791816
model-00004-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f283b790eaa5e7089f2183f17accec560af6e1de1730d712c6ae431c2ceb5a54
3
+ size 4954791848
model-00005-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c525ef96bc4f433a2c843d5e2ceb10829fd14a714f8846697cfb85184427c1e3
3
+ size 4954791848
model-00006-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e86dfd4ad461a8980c2e9c7c01c2c3214af56c366b166731c50a8509783200b
3
+ size 4954791848
model-00007-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f91c4c9d2f09374d433c3a03f10951f805ac9e4ece43e9922bf26fc5cba9be3
3
+ size 4954791848
model-00008-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6c998cd9f427f1c0947d08a0102068dc08cff4cc1643d313ddf8e8cfc399799
3
+ size 4954791848
model-00009-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beb51b9329c348e86163220b8928dd561ddd06ae27d2f51cb7607fd10af966a5
3
+ size 4954791848
model-00010-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec212cb30551ec5ded62b8648ccf70aec0d25057a0470cc4759748b2ae087206
3
+ size 4954791848
model-00011-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1035197cb7719f33feb4e5310795b156c3fd98c4137e14d1ec2e5d6076a0f270
3
+ size 4591469784
model-00012-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:118d8ba55296a64b8093ffca697debb197cca6428d5a7f9d2534cb3724bd2e23
3
+ size 2819260544
model.safetensors.index.json ADDED
@@ -0,0 +1,817 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 27009346304,
4
+ "total_size": 56837953024
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00012-of-00012.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00012.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00012.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00012.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00012.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00012.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00012.safetensors",
14
+ "model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00012.safetensors",
15
+ "model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00012.safetensors",
16
+ "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00012.safetensors",
17
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00012.safetensors",
18
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00012.safetensors",
19
+ "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00012.safetensors",
20
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00012.safetensors",
21
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00012.safetensors",
22
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00012.safetensors",
23
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00012.safetensors",
24
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00012.safetensors",
25
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00012.safetensors",
26
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00012.safetensors",
27
+ "model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00012.safetensors",
28
+ "model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00012.safetensors",
29
+ "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00012.safetensors",
30
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00012.safetensors",
31
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00012.safetensors",
32
+ "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00012.safetensors",
33
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00012.safetensors",
34
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00012.safetensors",
35
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00012.safetensors",
36
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
37
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
38
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
39
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
40
+ "model.layers.10.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
41
+ "model.layers.10.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
42
+ "model.layers.10.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
43
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
44
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
45
+ "model.layers.10.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
46
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
47
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
48
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00012.safetensors",
49
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
50
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
51
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
52
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
53
+ "model.layers.11.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
54
+ "model.layers.11.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
55
+ "model.layers.11.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
56
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
57
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
58
+ "model.layers.11.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
59
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
60
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
61
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00012.safetensors",
62
+ "model.layers.12.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
63
+ "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
64
+ "model.layers.12.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
65
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
66
+ "model.layers.12.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
67
+ "model.layers.12.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
68
+ "model.layers.12.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
69
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
70
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
71
+ "model.layers.12.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
72
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
73
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
74
+ "model.layers.13.input_layernorm.weight": "model-00003-of-00012.safetensors",
75
+ "model.layers.13.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
76
+ "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
77
+ "model.layers.13.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
78
+ "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
79
+ "model.layers.13.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
80
+ "model.layers.13.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
81
+ "model.layers.13.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
82
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
83
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
84
+ "model.layers.13.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
85
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
86
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
87
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00012.safetensors",
88
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
89
+ "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
90
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
91
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
92
+ "model.layers.14.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
93
+ "model.layers.14.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
94
+ "model.layers.14.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
95
+ "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
96
+ "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
97
+ "model.layers.14.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
98
+ "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
99
+ "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
100
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00012.safetensors",
101
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
102
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
103
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
104
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
105
+ "model.layers.15.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
106
+ "model.layers.15.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
107
+ "model.layers.15.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
108
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
109
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
110
+ "model.layers.15.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
111
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
112
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
113
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00012.safetensors",
114
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
115
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
116
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
117
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
118
+ "model.layers.16.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
119
+ "model.layers.16.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
120
+ "model.layers.16.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
121
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
122
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
123
+ "model.layers.16.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
124
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
125
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
126
+ "model.layers.17.input_layernorm.weight": "model-00004-of-00012.safetensors",
127
+ "model.layers.17.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
128
+ "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
129
+ "model.layers.17.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
130
+ "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
131
+ "model.layers.17.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
132
+ "model.layers.17.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
133
+ "model.layers.17.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
134
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
135
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
136
+ "model.layers.17.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
137
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
138
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
139
+ "model.layers.18.input_layernorm.weight": "model-00004-of-00012.safetensors",
140
+ "model.layers.18.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
141
+ "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
142
+ "model.layers.18.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
143
+ "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
144
+ "model.layers.18.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
145
+ "model.layers.18.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
146
+ "model.layers.18.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
147
+ "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
148
+ "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
149
+ "model.layers.18.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
150
+ "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
151
+ "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
152
+ "model.layers.19.input_layernorm.weight": "model-00004-of-00012.safetensors",
153
+ "model.layers.19.mlp.down_proj.weight": "model-00004-of-00012.safetensors",
154
+ "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
155
+ "model.layers.19.mlp.up_proj.weight": "model-00004-of-00012.safetensors",
156
+ "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00012.safetensors",
157
+ "model.layers.19.post_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
158
+ "model.layers.19.pre_feedforward_layernorm.weight": "model-00004-of-00012.safetensors",
159
+ "model.layers.19.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
160
+ "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
161
+ "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
162
+ "model.layers.19.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
163
+ "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
164
+ "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
165
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00012.safetensors",
166
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
167
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00012.safetensors",
168
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
169
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
170
+ "model.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
171
+ "model.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
172
+ "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00012.safetensors",
173
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00012.safetensors",
174
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00012.safetensors",
175
+ "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00012.safetensors",
176
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00012.safetensors",
177
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00012.safetensors",
178
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00012.safetensors",
179
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
180
+ "model.layers.20.mlp.gate_proj.weight": "model-00004-of-00012.safetensors",
181
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
182
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
183
+ "model.layers.20.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
184
+ "model.layers.20.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
185
+ "model.layers.20.self_attn.k_norm.weight": "model-00004-of-00012.safetensors",
186
+ "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00012.safetensors",
187
+ "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00012.safetensors",
188
+ "model.layers.20.self_attn.q_norm.weight": "model-00004-of-00012.safetensors",
189
+ "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00012.safetensors",
190
+ "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00012.safetensors",
191
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00012.safetensors",
192
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
193
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
194
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
195
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
196
+ "model.layers.21.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
197
+ "model.layers.21.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
198
+ "model.layers.21.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
199
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
200
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
201
+ "model.layers.21.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
202
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
203
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
204
+ "model.layers.22.input_layernorm.weight": "model-00005-of-00012.safetensors",
205
+ "model.layers.22.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
206
+ "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
207
+ "model.layers.22.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
208
+ "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
209
+ "model.layers.22.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
210
+ "model.layers.22.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
211
+ "model.layers.22.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
212
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
213
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
214
+ "model.layers.22.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
215
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
216
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
217
+ "model.layers.23.input_layernorm.weight": "model-00005-of-00012.safetensors",
218
+ "model.layers.23.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
219
+ "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
220
+ "model.layers.23.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
221
+ "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
222
+ "model.layers.23.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
223
+ "model.layers.23.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
224
+ "model.layers.23.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
225
+ "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
226
+ "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
227
+ "model.layers.23.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
228
+ "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
229
+ "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
230
+ "model.layers.24.input_layernorm.weight": "model-00005-of-00012.safetensors",
231
+ "model.layers.24.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
232
+ "model.layers.24.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
233
+ "model.layers.24.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
234
+ "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
235
+ "model.layers.24.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
236
+ "model.layers.24.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
237
+ "model.layers.24.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
238
+ "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
239
+ "model.layers.24.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
240
+ "model.layers.24.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
241
+ "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
242
+ "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
243
+ "model.layers.25.input_layernorm.weight": "model-00005-of-00012.safetensors",
244
+ "model.layers.25.mlp.down_proj.weight": "model-00005-of-00012.safetensors",
245
+ "model.layers.25.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
246
+ "model.layers.25.mlp.up_proj.weight": "model-00005-of-00012.safetensors",
247
+ "model.layers.25.post_attention_layernorm.weight": "model-00005-of-00012.safetensors",
248
+ "model.layers.25.post_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
249
+ "model.layers.25.pre_feedforward_layernorm.weight": "model-00005-of-00012.safetensors",
250
+ "model.layers.25.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
251
+ "model.layers.25.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
252
+ "model.layers.25.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
253
+ "model.layers.25.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
254
+ "model.layers.25.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
255
+ "model.layers.25.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
256
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00012.safetensors",
257
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
258
+ "model.layers.26.mlp.gate_proj.weight": "model-00005-of-00012.safetensors",
259
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
260
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
261
+ "model.layers.26.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
262
+ "model.layers.26.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
263
+ "model.layers.26.self_attn.k_norm.weight": "model-00005-of-00012.safetensors",
264
+ "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00012.safetensors",
265
+ "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00012.safetensors",
266
+ "model.layers.26.self_attn.q_norm.weight": "model-00005-of-00012.safetensors",
267
+ "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00012.safetensors",
268
+ "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00012.safetensors",
269
+ "model.layers.27.input_layernorm.weight": "model-00006-of-00012.safetensors",
270
+ "model.layers.27.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
271
+ "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
272
+ "model.layers.27.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
273
+ "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
274
+ "model.layers.27.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
275
+ "model.layers.27.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
276
+ "model.layers.27.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
277
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
278
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
279
+ "model.layers.27.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
280
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
281
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
282
+ "model.layers.28.input_layernorm.weight": "model-00006-of-00012.safetensors",
283
+ "model.layers.28.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
284
+ "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
285
+ "model.layers.28.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
286
+ "model.layers.28.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
287
+ "model.layers.28.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
288
+ "model.layers.28.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
289
+ "model.layers.28.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
290
+ "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
291
+ "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
292
+ "model.layers.28.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
293
+ "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
294
+ "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
295
+ "model.layers.29.input_layernorm.weight": "model-00006-of-00012.safetensors",
296
+ "model.layers.29.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
297
+ "model.layers.29.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
298
+ "model.layers.29.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
299
+ "model.layers.29.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
300
+ "model.layers.29.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
301
+ "model.layers.29.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
302
+ "model.layers.29.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
303
+ "model.layers.29.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
304
+ "model.layers.29.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
305
+ "model.layers.29.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
306
+ "model.layers.29.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
307
+ "model.layers.29.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
308
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00012.safetensors",
309
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
310
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
311
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
312
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
313
+ "model.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
314
+ "model.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
315
+ "model.layers.3.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
316
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
317
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
318
+ "model.layers.3.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
319
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
320
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
321
+ "model.layers.30.input_layernorm.weight": "model-00006-of-00012.safetensors",
322
+ "model.layers.30.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
323
+ "model.layers.30.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
324
+ "model.layers.30.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
325
+ "model.layers.30.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
326
+ "model.layers.30.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
327
+ "model.layers.30.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
328
+ "model.layers.30.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
329
+ "model.layers.30.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
330
+ "model.layers.30.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
331
+ "model.layers.30.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
332
+ "model.layers.30.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
333
+ "model.layers.30.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
334
+ "model.layers.31.input_layernorm.weight": "model-00006-of-00012.safetensors",
335
+ "model.layers.31.mlp.down_proj.weight": "model-00006-of-00012.safetensors",
336
+ "model.layers.31.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
337
+ "model.layers.31.mlp.up_proj.weight": "model-00006-of-00012.safetensors",
338
+ "model.layers.31.post_attention_layernorm.weight": "model-00006-of-00012.safetensors",
339
+ "model.layers.31.post_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
340
+ "model.layers.31.pre_feedforward_layernorm.weight": "model-00006-of-00012.safetensors",
341
+ "model.layers.31.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
342
+ "model.layers.31.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
343
+ "model.layers.31.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
344
+ "model.layers.31.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
345
+ "model.layers.31.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
346
+ "model.layers.31.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
347
+ "model.layers.32.input_layernorm.weight": "model-00007-of-00012.safetensors",
348
+ "model.layers.32.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
349
+ "model.layers.32.mlp.gate_proj.weight": "model-00006-of-00012.safetensors",
350
+ "model.layers.32.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
351
+ "model.layers.32.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
352
+ "model.layers.32.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
353
+ "model.layers.32.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
354
+ "model.layers.32.self_attn.k_norm.weight": "model-00006-of-00012.safetensors",
355
+ "model.layers.32.self_attn.k_proj.weight": "model-00006-of-00012.safetensors",
356
+ "model.layers.32.self_attn.o_proj.weight": "model-00006-of-00012.safetensors",
357
+ "model.layers.32.self_attn.q_norm.weight": "model-00006-of-00012.safetensors",
358
+ "model.layers.32.self_attn.q_proj.weight": "model-00006-of-00012.safetensors",
359
+ "model.layers.32.self_attn.v_proj.weight": "model-00006-of-00012.safetensors",
360
+ "model.layers.33.input_layernorm.weight": "model-00007-of-00012.safetensors",
361
+ "model.layers.33.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
362
+ "model.layers.33.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
363
+ "model.layers.33.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
364
+ "model.layers.33.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
365
+ "model.layers.33.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
366
+ "model.layers.33.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
367
+ "model.layers.33.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
368
+ "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
369
+ "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
370
+ "model.layers.33.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
371
+ "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
372
+ "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
373
+ "model.layers.34.input_layernorm.weight": "model-00007-of-00012.safetensors",
374
+ "model.layers.34.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
375
+ "model.layers.34.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
376
+ "model.layers.34.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
377
+ "model.layers.34.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
378
+ "model.layers.34.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
379
+ "model.layers.34.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
380
+ "model.layers.34.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
381
+ "model.layers.34.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
382
+ "model.layers.34.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
383
+ "model.layers.34.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
384
+ "model.layers.34.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
385
+ "model.layers.34.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
386
+ "model.layers.35.input_layernorm.weight": "model-00007-of-00012.safetensors",
387
+ "model.layers.35.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
388
+ "model.layers.35.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
389
+ "model.layers.35.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
390
+ "model.layers.35.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
391
+ "model.layers.35.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
392
+ "model.layers.35.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
393
+ "model.layers.35.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
394
+ "model.layers.35.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
395
+ "model.layers.35.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
396
+ "model.layers.35.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
397
+ "model.layers.35.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
398
+ "model.layers.35.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
399
+ "model.layers.36.input_layernorm.weight": "model-00007-of-00012.safetensors",
400
+ "model.layers.36.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
401
+ "model.layers.36.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
402
+ "model.layers.36.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
403
+ "model.layers.36.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
404
+ "model.layers.36.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
405
+ "model.layers.36.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
406
+ "model.layers.36.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
407
+ "model.layers.36.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
408
+ "model.layers.36.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
409
+ "model.layers.36.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
410
+ "model.layers.36.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
411
+ "model.layers.36.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
412
+ "model.layers.37.input_layernorm.weight": "model-00007-of-00012.safetensors",
413
+ "model.layers.37.mlp.down_proj.weight": "model-00007-of-00012.safetensors",
414
+ "model.layers.37.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
415
+ "model.layers.37.mlp.up_proj.weight": "model-00007-of-00012.safetensors",
416
+ "model.layers.37.post_attention_layernorm.weight": "model-00007-of-00012.safetensors",
417
+ "model.layers.37.post_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
418
+ "model.layers.37.pre_feedforward_layernorm.weight": "model-00007-of-00012.safetensors",
419
+ "model.layers.37.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
420
+ "model.layers.37.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
421
+ "model.layers.37.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
422
+ "model.layers.37.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
423
+ "model.layers.37.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
424
+ "model.layers.37.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
425
+ "model.layers.38.input_layernorm.weight": "model-00008-of-00012.safetensors",
426
+ "model.layers.38.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
427
+ "model.layers.38.mlp.gate_proj.weight": "model-00007-of-00012.safetensors",
428
+ "model.layers.38.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
429
+ "model.layers.38.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
430
+ "model.layers.38.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
431
+ "model.layers.38.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
432
+ "model.layers.38.self_attn.k_norm.weight": "model-00007-of-00012.safetensors",
433
+ "model.layers.38.self_attn.k_proj.weight": "model-00007-of-00012.safetensors",
434
+ "model.layers.38.self_attn.o_proj.weight": "model-00007-of-00012.safetensors",
435
+ "model.layers.38.self_attn.q_norm.weight": "model-00007-of-00012.safetensors",
436
+ "model.layers.38.self_attn.q_proj.weight": "model-00007-of-00012.safetensors",
437
+ "model.layers.38.self_attn.v_proj.weight": "model-00007-of-00012.safetensors",
438
+ "model.layers.39.input_layernorm.weight": "model-00008-of-00012.safetensors",
439
+ "model.layers.39.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
440
+ "model.layers.39.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
441
+ "model.layers.39.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
442
+ "model.layers.39.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
443
+ "model.layers.39.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
444
+ "model.layers.39.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
445
+ "model.layers.39.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
446
+ "model.layers.39.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
447
+ "model.layers.39.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
448
+ "model.layers.39.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
449
+ "model.layers.39.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
450
+ "model.layers.39.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
451
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00012.safetensors",
452
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
453
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
454
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
455
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
456
+ "model.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
457
+ "model.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
458
+ "model.layers.4.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
459
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
460
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
461
+ "model.layers.4.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
462
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
463
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
464
+ "model.layers.40.input_layernorm.weight": "model-00008-of-00012.safetensors",
465
+ "model.layers.40.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
466
+ "model.layers.40.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
467
+ "model.layers.40.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
468
+ "model.layers.40.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
469
+ "model.layers.40.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
470
+ "model.layers.40.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
471
+ "model.layers.40.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
472
+ "model.layers.40.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
473
+ "model.layers.40.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
474
+ "model.layers.40.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
475
+ "model.layers.40.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
476
+ "model.layers.40.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
477
+ "model.layers.41.input_layernorm.weight": "model-00008-of-00012.safetensors",
478
+ "model.layers.41.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
479
+ "model.layers.41.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
480
+ "model.layers.41.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
481
+ "model.layers.41.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
482
+ "model.layers.41.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
483
+ "model.layers.41.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
484
+ "model.layers.41.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
485
+ "model.layers.41.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
486
+ "model.layers.41.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
487
+ "model.layers.41.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
488
+ "model.layers.41.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
489
+ "model.layers.41.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
490
+ "model.layers.42.input_layernorm.weight": "model-00008-of-00012.safetensors",
491
+ "model.layers.42.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
492
+ "model.layers.42.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
493
+ "model.layers.42.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
494
+ "model.layers.42.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
495
+ "model.layers.42.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
496
+ "model.layers.42.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
497
+ "model.layers.42.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
498
+ "model.layers.42.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
499
+ "model.layers.42.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
500
+ "model.layers.42.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
501
+ "model.layers.42.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
502
+ "model.layers.42.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
503
+ "model.layers.43.input_layernorm.weight": "model-00008-of-00012.safetensors",
504
+ "model.layers.43.mlp.down_proj.weight": "model-00008-of-00012.safetensors",
505
+ "model.layers.43.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
506
+ "model.layers.43.mlp.up_proj.weight": "model-00008-of-00012.safetensors",
507
+ "model.layers.43.post_attention_layernorm.weight": "model-00008-of-00012.safetensors",
508
+ "model.layers.43.post_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
509
+ "model.layers.43.pre_feedforward_layernorm.weight": "model-00008-of-00012.safetensors",
510
+ "model.layers.43.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
511
+ "model.layers.43.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
512
+ "model.layers.43.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
513
+ "model.layers.43.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
514
+ "model.layers.43.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
515
+ "model.layers.43.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
516
+ "model.layers.44.input_layernorm.weight": "model-00009-of-00012.safetensors",
517
+ "model.layers.44.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
518
+ "model.layers.44.mlp.gate_proj.weight": "model-00008-of-00012.safetensors",
519
+ "model.layers.44.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
520
+ "model.layers.44.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
521
+ "model.layers.44.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
522
+ "model.layers.44.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
523
+ "model.layers.44.self_attn.k_norm.weight": "model-00008-of-00012.safetensors",
524
+ "model.layers.44.self_attn.k_proj.weight": "model-00008-of-00012.safetensors",
525
+ "model.layers.44.self_attn.o_proj.weight": "model-00008-of-00012.safetensors",
526
+ "model.layers.44.self_attn.q_norm.weight": "model-00008-of-00012.safetensors",
527
+ "model.layers.44.self_attn.q_proj.weight": "model-00008-of-00012.safetensors",
528
+ "model.layers.44.self_attn.v_proj.weight": "model-00008-of-00012.safetensors",
529
+ "model.layers.45.input_layernorm.weight": "model-00009-of-00012.safetensors",
530
+ "model.layers.45.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
531
+ "model.layers.45.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
532
+ "model.layers.45.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
533
+ "model.layers.45.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
534
+ "model.layers.45.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
535
+ "model.layers.45.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
536
+ "model.layers.45.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
537
+ "model.layers.45.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
538
+ "model.layers.45.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
539
+ "model.layers.45.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
540
+ "model.layers.45.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
541
+ "model.layers.45.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
542
+ "model.layers.46.input_layernorm.weight": "model-00009-of-00012.safetensors",
543
+ "model.layers.46.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
544
+ "model.layers.46.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
545
+ "model.layers.46.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
546
+ "model.layers.46.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
547
+ "model.layers.46.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
548
+ "model.layers.46.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
549
+ "model.layers.46.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
550
+ "model.layers.46.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
551
+ "model.layers.46.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
552
+ "model.layers.46.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
553
+ "model.layers.46.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
554
+ "model.layers.46.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
555
+ "model.layers.47.input_layernorm.weight": "model-00009-of-00012.safetensors",
556
+ "model.layers.47.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
557
+ "model.layers.47.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
558
+ "model.layers.47.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
559
+ "model.layers.47.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
560
+ "model.layers.47.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
561
+ "model.layers.47.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
562
+ "model.layers.47.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
563
+ "model.layers.47.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
564
+ "model.layers.47.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
565
+ "model.layers.47.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
566
+ "model.layers.47.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
567
+ "model.layers.47.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
568
+ "model.layers.48.input_layernorm.weight": "model-00009-of-00012.safetensors",
569
+ "model.layers.48.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
570
+ "model.layers.48.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
571
+ "model.layers.48.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
572
+ "model.layers.48.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
573
+ "model.layers.48.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
574
+ "model.layers.48.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
575
+ "model.layers.48.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
576
+ "model.layers.48.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
577
+ "model.layers.48.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
578
+ "model.layers.48.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
579
+ "model.layers.48.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
580
+ "model.layers.48.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
581
+ "model.layers.49.input_layernorm.weight": "model-00009-of-00012.safetensors",
582
+ "model.layers.49.mlp.down_proj.weight": "model-00009-of-00012.safetensors",
583
+ "model.layers.49.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
584
+ "model.layers.49.mlp.up_proj.weight": "model-00009-of-00012.safetensors",
585
+ "model.layers.49.post_attention_layernorm.weight": "model-00009-of-00012.safetensors",
586
+ "model.layers.49.post_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
587
+ "model.layers.49.pre_feedforward_layernorm.weight": "model-00009-of-00012.safetensors",
588
+ "model.layers.49.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
589
+ "model.layers.49.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
590
+ "model.layers.49.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
591
+ "model.layers.49.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
592
+ "model.layers.49.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
593
+ "model.layers.49.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
594
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00012.safetensors",
595
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
596
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
597
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
598
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
599
+ "model.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
600
+ "model.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
601
+ "model.layers.5.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
602
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
603
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
604
+ "model.layers.5.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
605
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
606
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
607
+ "model.layers.50.input_layernorm.weight": "model-00010-of-00012.safetensors",
608
+ "model.layers.50.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
609
+ "model.layers.50.mlp.gate_proj.weight": "model-00009-of-00012.safetensors",
610
+ "model.layers.50.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
611
+ "model.layers.50.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
612
+ "model.layers.50.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
613
+ "model.layers.50.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
614
+ "model.layers.50.self_attn.k_norm.weight": "model-00009-of-00012.safetensors",
615
+ "model.layers.50.self_attn.k_proj.weight": "model-00009-of-00012.safetensors",
616
+ "model.layers.50.self_attn.o_proj.weight": "model-00009-of-00012.safetensors",
617
+ "model.layers.50.self_attn.q_norm.weight": "model-00009-of-00012.safetensors",
618
+ "model.layers.50.self_attn.q_proj.weight": "model-00009-of-00012.safetensors",
619
+ "model.layers.50.self_attn.v_proj.weight": "model-00009-of-00012.safetensors",
620
+ "model.layers.51.input_layernorm.weight": "model-00010-of-00012.safetensors",
621
+ "model.layers.51.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
622
+ "model.layers.51.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
623
+ "model.layers.51.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
624
+ "model.layers.51.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
625
+ "model.layers.51.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
626
+ "model.layers.51.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
627
+ "model.layers.51.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
628
+ "model.layers.51.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
629
+ "model.layers.51.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
630
+ "model.layers.51.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
631
+ "model.layers.51.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
632
+ "model.layers.51.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
633
+ "model.layers.52.input_layernorm.weight": "model-00010-of-00012.safetensors",
634
+ "model.layers.52.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
635
+ "model.layers.52.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
636
+ "model.layers.52.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
637
+ "model.layers.52.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
638
+ "model.layers.52.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
639
+ "model.layers.52.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
640
+ "model.layers.52.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
641
+ "model.layers.52.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
642
+ "model.layers.52.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
643
+ "model.layers.52.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
644
+ "model.layers.52.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
645
+ "model.layers.52.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
646
+ "model.layers.53.input_layernorm.weight": "model-00010-of-00012.safetensors",
647
+ "model.layers.53.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
648
+ "model.layers.53.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
649
+ "model.layers.53.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
650
+ "model.layers.53.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
651
+ "model.layers.53.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
652
+ "model.layers.53.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
653
+ "model.layers.53.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
654
+ "model.layers.53.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
655
+ "model.layers.53.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
656
+ "model.layers.53.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
657
+ "model.layers.53.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
658
+ "model.layers.53.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
659
+ "model.layers.54.input_layernorm.weight": "model-00010-of-00012.safetensors",
660
+ "model.layers.54.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
661
+ "model.layers.54.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
662
+ "model.layers.54.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
663
+ "model.layers.54.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
664
+ "model.layers.54.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
665
+ "model.layers.54.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
666
+ "model.layers.54.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
667
+ "model.layers.54.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
668
+ "model.layers.54.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
669
+ "model.layers.54.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
670
+ "model.layers.54.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
671
+ "model.layers.54.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
672
+ "model.layers.55.input_layernorm.weight": "model-00010-of-00012.safetensors",
673
+ "model.layers.55.mlp.down_proj.weight": "model-00010-of-00012.safetensors",
674
+ "model.layers.55.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
675
+ "model.layers.55.mlp.up_proj.weight": "model-00010-of-00012.safetensors",
676
+ "model.layers.55.post_attention_layernorm.weight": "model-00010-of-00012.safetensors",
677
+ "model.layers.55.post_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
678
+ "model.layers.55.pre_feedforward_layernorm.weight": "model-00010-of-00012.safetensors",
679
+ "model.layers.55.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
680
+ "model.layers.55.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
681
+ "model.layers.55.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
682
+ "model.layers.55.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
683
+ "model.layers.55.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
684
+ "model.layers.55.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
685
+ "model.layers.56.input_layernorm.weight": "model-00011-of-00012.safetensors",
686
+ "model.layers.56.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
687
+ "model.layers.56.mlp.gate_proj.weight": "model-00010-of-00012.safetensors",
688
+ "model.layers.56.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
689
+ "model.layers.56.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
690
+ "model.layers.56.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
691
+ "model.layers.56.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
692
+ "model.layers.56.self_attn.k_norm.weight": "model-00010-of-00012.safetensors",
693
+ "model.layers.56.self_attn.k_proj.weight": "model-00010-of-00012.safetensors",
694
+ "model.layers.56.self_attn.o_proj.weight": "model-00010-of-00012.safetensors",
695
+ "model.layers.56.self_attn.q_norm.weight": "model-00010-of-00012.safetensors",
696
+ "model.layers.56.self_attn.q_proj.weight": "model-00010-of-00012.safetensors",
697
+ "model.layers.56.self_attn.v_proj.weight": "model-00010-of-00012.safetensors",
698
+ "model.layers.57.input_layernorm.weight": "model-00011-of-00012.safetensors",
699
+ "model.layers.57.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
700
+ "model.layers.57.mlp.gate_proj.weight": "model-00011-of-00012.safetensors",
701
+ "model.layers.57.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
702
+ "model.layers.57.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
703
+ "model.layers.57.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
704
+ "model.layers.57.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
705
+ "model.layers.57.self_attn.k_norm.weight": "model-00011-of-00012.safetensors",
706
+ "model.layers.57.self_attn.k_proj.weight": "model-00011-of-00012.safetensors",
707
+ "model.layers.57.self_attn.o_proj.weight": "model-00011-of-00012.safetensors",
708
+ "model.layers.57.self_attn.q_norm.weight": "model-00011-of-00012.safetensors",
709
+ "model.layers.57.self_attn.q_proj.weight": "model-00011-of-00012.safetensors",
710
+ "model.layers.57.self_attn.v_proj.weight": "model-00011-of-00012.safetensors",
711
+ "model.layers.58.input_layernorm.weight": "model-00011-of-00012.safetensors",
712
+ "model.layers.58.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
713
+ "model.layers.58.mlp.gate_proj.weight": "model-00011-of-00012.safetensors",
714
+ "model.layers.58.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
715
+ "model.layers.58.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
716
+ "model.layers.58.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
717
+ "model.layers.58.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
718
+ "model.layers.58.self_attn.k_norm.weight": "model-00011-of-00012.safetensors",
719
+ "model.layers.58.self_attn.k_proj.weight": "model-00011-of-00012.safetensors",
720
+ "model.layers.58.self_attn.o_proj.weight": "model-00011-of-00012.safetensors",
721
+ "model.layers.58.self_attn.q_norm.weight": "model-00011-of-00012.safetensors",
722
+ "model.layers.58.self_attn.q_proj.weight": "model-00011-of-00012.safetensors",
723
+ "model.layers.58.self_attn.v_proj.weight": "model-00011-of-00012.safetensors",
724
+ "model.layers.59.input_layernorm.weight": "model-00011-of-00012.safetensors",
725
+ "model.layers.59.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
726
+ "model.layers.59.mlp.gate_proj.weight": "model-00011-of-00012.safetensors",
727
+ "model.layers.59.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
728
+ "model.layers.59.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
729
+ "model.layers.59.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
730
+ "model.layers.59.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
731
+ "model.layers.59.self_attn.k_norm.weight": "model-00011-of-00012.safetensors",
732
+ "model.layers.59.self_attn.k_proj.weight": "model-00011-of-00012.safetensors",
733
+ "model.layers.59.self_attn.o_proj.weight": "model-00011-of-00012.safetensors",
734
+ "model.layers.59.self_attn.q_norm.weight": "model-00011-of-00012.safetensors",
735
+ "model.layers.59.self_attn.q_proj.weight": "model-00011-of-00012.safetensors",
736
+ "model.layers.59.self_attn.v_proj.weight": "model-00011-of-00012.safetensors",
737
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00012.safetensors",
738
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
739
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
740
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
741
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
742
+ "model.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
743
+ "model.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
744
+ "model.layers.6.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
745
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
746
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
747
+ "model.layers.6.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
748
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
749
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
750
+ "model.layers.60.input_layernorm.weight": "model-00011-of-00012.safetensors",
751
+ "model.layers.60.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
752
+ "model.layers.60.mlp.gate_proj.weight": "model-00011-of-00012.safetensors",
753
+ "model.layers.60.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
754
+ "model.layers.60.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
755
+ "model.layers.60.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
756
+ "model.layers.60.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
757
+ "model.layers.60.self_attn.k_norm.weight": "model-00011-of-00012.safetensors",
758
+ "model.layers.60.self_attn.k_proj.weight": "model-00011-of-00012.safetensors",
759
+ "model.layers.60.self_attn.o_proj.weight": "model-00011-of-00012.safetensors",
760
+ "model.layers.60.self_attn.q_norm.weight": "model-00011-of-00012.safetensors",
761
+ "model.layers.60.self_attn.q_proj.weight": "model-00011-of-00012.safetensors",
762
+ "model.layers.60.self_attn.v_proj.weight": "model-00011-of-00012.safetensors",
763
+ "model.layers.61.input_layernorm.weight": "model-00011-of-00012.safetensors",
764
+ "model.layers.61.mlp.down_proj.weight": "model-00011-of-00012.safetensors",
765
+ "model.layers.61.mlp.gate_proj.weight": "model-00011-of-00012.safetensors",
766
+ "model.layers.61.mlp.up_proj.weight": "model-00011-of-00012.safetensors",
767
+ "model.layers.61.post_attention_layernorm.weight": "model-00011-of-00012.safetensors",
768
+ "model.layers.61.post_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
769
+ "model.layers.61.pre_feedforward_layernorm.weight": "model-00011-of-00012.safetensors",
770
+ "model.layers.61.self_attn.k_norm.weight": "model-00011-of-00012.safetensors",
771
+ "model.layers.61.self_attn.k_proj.weight": "model-00011-of-00012.safetensors",
772
+ "model.layers.61.self_attn.o_proj.weight": "model-00011-of-00012.safetensors",
773
+ "model.layers.61.self_attn.q_norm.weight": "model-00011-of-00012.safetensors",
774
+ "model.layers.61.self_attn.q_proj.weight": "model-00011-of-00012.safetensors",
775
+ "model.layers.61.self_attn.v_proj.weight": "model-00011-of-00012.safetensors",
776
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00012.safetensors",
777
+ "model.layers.7.mlp.down_proj.weight": "model-00002-of-00012.safetensors",
778
+ "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
779
+ "model.layers.7.mlp.up_proj.weight": "model-00002-of-00012.safetensors",
780
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00012.safetensors",
781
+ "model.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
782
+ "model.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00012.safetensors",
783
+ "model.layers.7.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
784
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
785
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
786
+ "model.layers.7.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
787
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
788
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
789
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00012.safetensors",
790
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
791
+ "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00012.safetensors",
792
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
793
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
794
+ "model.layers.8.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
795
+ "model.layers.8.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
796
+ "model.layers.8.self_attn.k_norm.weight": "model-00002-of-00012.safetensors",
797
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00012.safetensors",
798
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00012.safetensors",
799
+ "model.layers.8.self_attn.q_norm.weight": "model-00002-of-00012.safetensors",
800
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00012.safetensors",
801
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00012.safetensors",
802
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00012.safetensors",
803
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00012.safetensors",
804
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00012.safetensors",
805
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00012.safetensors",
806
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00012.safetensors",
807
+ "model.layers.9.post_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
808
+ "model.layers.9.pre_feedforward_layernorm.weight": "model-00003-of-00012.safetensors",
809
+ "model.layers.9.self_attn.k_norm.weight": "model-00003-of-00012.safetensors",
810
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00012.safetensors",
811
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00012.safetensors",
812
+ "model.layers.9.self_attn.q_norm.weight": "model-00003-of-00012.safetensors",
813
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00012.safetensors",
814
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00012.safetensors",
815
+ "model.norm.weight": "model-00011-of-00012.safetensors"
816
+ }
817
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<bos>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<end_of_turn>",
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": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
wmt24pp_12b.md ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ WMT24++ ChrF++ Metrics - Updated with YanoljaNEXT-Rosetta-12B-2510
2
+
3
+ | Model | Avg | ar_EG | ar_SA | bg_BG | bn_IN | ca_ES | cs_CZ | da_DK | de_DE | el_GR | es_MX | et_EE | fa_IR | fi_FI | fil_PH | fr_CA | fr_FR | gu_IN | he_IL | hi_IN | hr_HR | hu_HU | id_ID | is_IS | it_IT | ja_JP | kn_IN |
4
+ | ----------------------- | ---- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ------ | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
5
+ | DeepL Translate | 54.9 | 33.4 | 37.8 | 61.8 | | | 55.7 | 62.1 | 60.5 | 61.5 | 67.7 | 57.5 | | 62.0 | | 64.2 | 61.5 | | | 53.3 | 60.9 | | 66.0 | 34.9 | | | |
6
+ | YanoljaNEXT-Rosetta-12B | 54.4 | 32.8 | 37.9 | 62.0 | 40.7 | 60.3 | 55.3 | 65.9 | 58.1 | 62.3 | 67.5 | 50.2 | 54.0 | 59.6 | 61.5 | 68.3 | 62.7 | 51.8 | 54.1 | 39.4 | | 53.2 | 61.5 | | 65.0 | 32.1 | |
7
+ | GPT-4o | 54.4 | 35.4 | 38.4 | 62.7 | 45.0 | 64.2 | 55.8 | 63.0 | 60.8 | 62.5 | 65.7 | 60.4 | 53.0 | 63.1 | 57.8 | 69.1 | 63.7 | 47.7 | 54.9 | 41.5 | 59.4 | 54.9 | 61.9 | 49.9 | 65.0 | 36.1 | 54.8 |
8
+ | Google Translate | 53.9 | 35.0 | 39.3 | 62.2 | 43.8 | 62.4 | 53.7 | 59.9 | 60.9 | 60.5 | 66.8 | 58.6 | 48.5 | 60.3 | 59.0 | 66.5 | 62.0 | 49.8 | 53.1 | 41.1 | 57.1 | 52.2 | 61.5 | 47.2 | 64.9 | 35.3 | 59.0 |
9
+ | YanoljaNEXT-Rosetta-4B | 53.6 | 32.4 | 37.3 | 61.2 | | | 54.3 | 65.1 | 56.9 | 61.6 | 66.4 | | 53.3 | 58.5 | 61.2 | 66.3 | 61.4 | 51.2 | 53.2 | 39.1 | | 52.1 | 60.9 | | 63.7 | 31.4 | |
10
+ | Claude-3.5 | 52.7 | 38.9 | 36.5 | 59.4 | 42.6 | 63.8 | 53.8 | 60.7 | 58.7 | 60.8 | 64.3 | 57.4 | 49.3 | 60.4 | 55.2 | 65.9 | 62.4 | 47.5 | 52.6 | 40.9 | 57.5 | 52.7 | 60.0 | 48.8 | 63.4 | 32.8 | 53.9 |
11
+ | Microsoft Translator | 52.6 | 33.4 | 37.3 | 61.6 | 43.7 | 63.0 | 54.5 | 61.1 | 59.0 | 59.8 | 65.2 | 57.6 | 49.3 | 61.0 | 55.1 | 61.8 | 58.2 | 49.1 | 54.0 | 37.5 | 57.3 | 51.5 | 59.0 | 47.7 | 64.7 | 32.8 | 60.9 |
12
+ | OpenAI o1 | 52.4 | 35.2 | 36.5 | 60.3 | 41.9 | 62.3 | 54.1 | 61.3 | 59.5 | 60.9 | 65.0 | 58.7 | 51.1 | 61.7 | 56.2 | 65.7 | 60.4 | 40.4 | 53.8 | 41.1 | 58.4 | 53.5 | 59.9 | 49.5 | 63.2 | 34.9 | 45.6 |
13
+ | Unbabel Tower | 52.0 | | | | | | 53.3 | | 59.4 | | 65.3 | | | | | 66.0 | 61.8 | | | 40.3 | | | | 47.9 | 64.8 | 33.1 | |
14
+ | Gemini-1.5-Pro | 52.0 | 40.0 | 36.7 | 58.6 | 41.7 | 63.4 | 53.4 | 62.6 | 58.8 | 58.3 | 63.8 | 56.7 | 47.4 | 58.1 | 53.4 | 64.0 | 60.9 | 46.8 | 52.3 | 41.3 | 56.4 | 51.5 | 58.7 | 47.6 | 62.9 | 33.8 | 52.4 |
15
+ | OpenAI o1-mini | 50.9 | 36.2 | 36.9 | 58.1 | 41.5 | 61.2 | 51.5 | 59.4 | 58.9 | 56.5 | 64.6 | 54.3 | 50.7 | 58.4 | 55.7 | 65.5 | 61.2 | 43.2 | 49.3 | 39.0 | 55.1 | 49.9 | 59.4 | 41.8 | 63.0 | 33.8 | 49.9 |
16
+ | Gemini-1.5-Flash | 50.4 | 33.4 | 35.7 | 56.8 | 42.1 | 60.6 | 51.4 | 59.2 | 57.8 | 56.4 | 64.2 | 53.9 | 47.5 | 56.5 | 54.6 | 62.9 | 59.4 | 45.1 | 49.4 | 39.9 | 54.4 | 49.8 | 58.6 | 43.4 | 61.8 | 32.6 | 53.0 |
17
+ | Yandex Translate | 49.2 | 30.3 | 34.0 | 57.4 | 38.8 | 58.8 | 51.2 | 55.9 | 57.1 | 58.2 | 64.5 | 54.9 | 42.5 | 59.0 | 56.9 | 64.1 | 60.2 | 44.4 | 49.5 | 35.3 | 50.0 | 49.6 | 59.2 | 43.5 | 61.6 | 25.5 | 55.4 |
18
+ | CommandR-plus | 48.7 | 32.1 | 35.7 | | | 49.9 | | 56.7 | 56.1 | 62.2 | | 46.0 | | | 63.0 | 59.4 | | 49.2 | 37.6 | | | 57.3 | | 61.5 | 29.8 | | |
19
+ | Aya23 | 48.1 | 31.3 | 35.2 | | | 48.7 | | 55.4 | 55.1 | 62.4 | | 47.4 | | | 62.8 | 58.3 | | 47.6 | 38.2 | | | 57.4 | | 60.7 | 28.9 | | |
20
+
21
+ | Model | ko_KR | lt_LT | lv_LV | ml_IN | mr_IN | nl_NL | no_NO | pa_IN | pl_PL | pt_BR | pt_PT | ro_RO | ru_RU | sk_SK | sl_SI | sr_RS | sv_SE | sw_KE | sw_TZ | ta_IN | te_IN | th_TH | tr_TR | uk_UA | ur_PK | vi_VN | zh_CN | zh_TW | zu_ZA |
22
+ | ----------------------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
23
+ | DeepL Translate | 37.0 | 53.0 | 55.2 | | | 60.6 | 67.8 | | 50.4 | 64.7 | 62.2 | 61.1 | 48.4 | 52.2 | 56.4 | | 65.6 | | | | | 60.3 | 61.5 | | | 37.4 | 18.3 | | |
24
+ | YanoljaNEXT-Rosetta-12B | 37.1 | | 51.4 | | | 60.1 | 65.2 | | 49.1 | 63.4 | 57.1 | 63.2 | 50.1 | 51.2 | | | 64.7 | | | | | 44.6 | 57.7 | 56.6 | | 57.0 | 35.5 | | |
25
+ | GPT-4o | 36.4 | 52.5 | 56.8 | 47.7 | 44.9 | 61.3 | 69.0 | 48.6 | 51.0 | 65.7 | 61.6 | 62.8 | 50.3 | 53.0 | 57.0 | 54.3 | 66.3 | 54.7 | 58.6 | 47.3 | 47.0 | 50.5 | 59.3 | 54.5 | 50.6 | 54.4 | 41.0 | 33.6 | 56.1 |
26
+ | Google Translate | 36.5 | 50.6 | 56.1 | 52.9 | 47.8 | 60.5 | 66.7 | 56.0 | 50.9 | 65.7 | 55.7 | 60.6 | 48.9 | 50.7 | 57.7 | 50.6 | 63.4 | 51.6 | 56.3 | 50.1 | 49.1 | 50.4 | 58.9 | 54.1 | 51.8 | 54.2 | 39.9 | 37.4 | 59.5 |
27
+ | YanoljaNEXT-Rosetta-4B | 35.1 | | | | | 59.3 | | | 48.3 | 62.4 | 56.6 | 62.9 | 49.4 | 50.9 | | | 64.4 | | | | | 44.2 | 56.5 | 55.5 | | 56.2 | 33.7 | | |
28
+ | Claude-3.5 | 33.7 | 50.4 | 55.9 | 49.6 | 43.9 | 60.0 | 65.2 | 50.7 | 49.6 | 64.4 | 61.3 | 60.8 | 50.0 | 51.6 | 55.0 | 53.2 | 64.3 | 52.9 | 55.1 | 47.8 | 46.0 | 48.4 | 57.8 | 53.0 | 47.1 | 51.0 | 36.1 | 28.9 | 52.1 |
29
+ | Microsoft Translator | 35.9 | 49.7 | 54.1 | 53.0 | 45.8 | 59.8 | 68.8 | 53.6 | 51.4 | 64.0 | 56.0 | 61.3 | 47.8 | 51.2 | 57.2 | 49.7 | 64.4 | 48.4 | 53.1 | 50.1 | 48.1 | 47.9 | 58.8 | 53.2 | 51.4 | 50.5 | 34.8 | 21.2 | 57.5 |
30
+ | OpenAI o1 | 33.6 | 51.5 | 55.4 | 43.9 | 44.3 | 60.8 | 67.1 | 44.2 | 49.7 | 64.2 | 60.6 | 61.5 | 48.8 | 51.9 | 55.7 | 53.6 | 64.3 | 53.1 | 56.3 | 45.2 | 43.4 | 49.5 | 56.8 | 54.0 | 48.4 | 52.2 | 37.6 | 32.1 | 53.1 |
31
+ | Unbabel Tower | 34.6 | | | | | 60.3 | | | | 64.8 | 59.8 | | 48.4 | | | | | | | | | | 53.6 | | | 38.4 | 32.4 | | |
32
+ | Gemini-1.5-Pro | 33.3 | 50.7 | 54.1 | 48.8 | 43.0 | 60.1 | 64.0 | 48.8 | 48.6 | 63.3 | 60.5 | 59.2 | 48.4 | 51.6 | 54.1 | 53.1 | 61.7 | 51.5 | 54.7 | 42.4 | 45.0 | 46.8 | 57.7 | 53.0 | 45.6 | 49.7 | 43.0 | 29.2 | 54.2 |
33
+ | OpenAI o1-mini | 34.1 | 47.1 | 50.3 | 43.0 | 42.2 | 58.6 | 65.8 | 46.3 | 48.8 | 64.0 | 58.4 | 59.6 | 47.7 | 49.0 | 53.2 | 48.8 | 63.2 | 49.2 | 52.7 | 44.2 | 43.7 | 46.3 | 55.9 | 51.7 | 48.4 | 50.9 | 36.6 | 32.1 | 51.6 |
34
+ | Gemini-1.5-Flash | 32.2 | 48.9 | 52.7 | 45.6 | 42.2 | 58.0 | 63.2 | 46.2 | 47.5 | 62.5 | 58.8 | 59.1 | 47.0 | 49.7 | 51.5 | 50.0 | 60.7 | 50.2 | 53.3 | 43.8 | 44.6 | 45.9 | 55.7 | 51.7 | 45.8 | 49.7 | 38.2 | 29.3 | 49.7 |
35
+ | Yandex Translate | 26.9 | 47.7 | 51.2 | 44.7 | 42.7 | 56.6 | 63.7 | 46.7 | 47.1 | | 56.2 | 58.4 | 48.4 | 48.4 | 54.3 | 46.3 | 60.9 | 50.2 | 53.3 | 49.1 | 42.3 | 35.7 | 56.6 | 54.1 | 48.5 | 47.3 | 29.3 | 18.7 | 57.0 |
36
+ | CommandR-plus | 30.3 | | | | | 57.3 | | | 46.7 | 62.5 | 59.0 | 57.6 | 45.7 | | | | | | | | | | 51.7 | 49.9 | | 48.5 | 33.5 | 27.3 | |
37
+ | Aya23 | 29.2 | | | | | 56.1 | | | 45.4 | 62.3 | 56.8 | 56.9 | 45.0 | | | | | | | | | 51.7 | 48.8 | | 48.9 | 31.7 | 28.3 | |
38
+
39
+