Hugging Face's logo

nexsendev
/
webicoder-v3-mlx-q8

Text Generation
MLX
Safetensors
English
phi
phi-2
html
css
web-development
code-generation
fine-tuned
apple-silicon
conversational
8-bit precision
Model card Files
xet
 Community
nexsendev commited on
Commit
fea7f6b
·
verified ·
1 Parent(s): 5fa60ac

Add files using upload-large-folder tool

Browse files
Files changed (16) hide show
  1. .cache/._huggingface +0 -0
  2. .gitattributes +1 -0
  3. LICENSE +21 -0
  4. README.md +307 -0
  5. added_tokens.json +40 -0
  6. config.json +43 -0
  7. example.py +198 -0
  8. generation_config.json +18 -0
  9. merges.txt +0 -0
  10. model.safetensors +3 -0
  11. model.safetensors.index.json +849 -0
  12. special_tokens_map.json +23 -0
  13. tokenizer.json +0 -0
  14. tokenizer_config.json +328 -0
  15. vocab.json +0 -0
  16. webicoder_icon.png +3 -0
.cache/._huggingface ADDED
Binary file (4.1 kB). View file
 
.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
+ webicoder_icon.png filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ MIT License
2
+
3
+ Copyright (c) 2025 WebICoder
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
README.md ADDED
@@ -0,0 +1,307 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - en
5
+ tags:
6
+ - mlx
7
+ - phi-2
8
+ - html
9
+ - css
10
+ - web-development
11
+ - code-generation
12
+ - fine-tuned
13
+ - apple-silicon
14
+ base_model: microsoft/phi-2
15
+ pipeline_tag: text-generation
16
+ library_name: mlx
17
+ model-index:
18
+ - name: WebICoder-v3-MLX-8bit
19
+ results: []
20
+ ---
21
+
22
+ # ⚡ WebICoder v3 — HTML Code Generation (MLX 8-bit)
23
+
24
+ **WebICoder v3** is a fine-tuned version of [Microsoft Phi-2](https://huggingface.co/microsoft/phi-2) (2.7B parameters) specialized in generating **complete, production-ready HTML/CSS websites** from natural language descriptions.
25
+
26
+ Optimized for **Apple Silicon** via [MLX](https://github.com/ml-explore/mlx).
27
+
28
+ ## Model Details
29
+
30
+ | Property | Value |
31
+ |---|---|
32
+ | **Base Model** | Microsoft Phi-2 (2.7B parameters) |
33
+ | **Architecture** | PhiForCausalLM (32 layers, 2560 hidden) |
34
+ | **Format** | MLX (Apple Silicon optimized) |
35
+ | **Quantization** | 8-bit (8.503 bits/weight, affine) |
36
+ | **Size** | ~2.9 GB |
37
+ | **Context Length** | 4096 tokens |
38
+ | **Task** | HTML/CSS Code Generation |
39
+ | **Speed** | ~12-20 tok/s on M-series Mac |
40
+
41
+ ## Also Available
42
+
43
+ | Variant | Link | Size |
44
+ |---|---|---|
45
+ | **8-bit** (higher quality) | `YOUR_USERNAME/WebICoder-v3-MLX-8bit` | ~2.9 GB |
46
+
47
+ ---
48
+
49
+ ## ⚠️ MANDATORY — Read Before Using
50
+
51
+ > **If you skip these steps, the model will produce broken, repeated, or low-quality output.**
52
+ > Follow ALL 5 rules below to get the best results.
53
+
54
+ ### Rule 1 — Use the correct prompt format
55
+
56
+ The model was trained with an **Alpaca-style format**. You MUST wrap your prompt like this:
57
+
58
+ ```
59
+ ### Instruction:
60
+ {your website description here}
61
+
62
+ ### Response:
63
+ ```
64
+
65
+ ❌ **DO NOT** send raw text like `"Create a website"` — the model won't understand it correctly.
66
+
67
+ ✅ **DO** use the format above, or use `tokenizer.apply_chat_template()` which does it automatically.
68
+
69
+ ### Rule 2 — ALWAYS stop at `</html>`
70
+
71
+ The model does not always emit an EOS token after finishing the HTML. You **MUST** check for `</html>` in the output and stop generation when you see it.
72
+
73
+ ```python
74
+ # ✅ Correct — stop at </html>
75
+ for response in stream_generate(model, tokenizer, prompt=prompt, max_tokens=4096, sampler=sampler):
76
+ full_text += response.text
77
+ if "</html>" in full_text:
78
+ break
79
+ ```
80
+
81
+ ❌ Without this, the model will **repeat the entire page** in a loop.
82
+
83
+ ### Rule 3 — Use repetition penalty
84
+
85
+ A repetition penalty is **essential** to prevent the model from generating duplicate sections (e.g., the same footer twice, identical testimonials).
86
+
87
+ ```python
88
+ from mlx_lm.sample_utils import make_logits_processors
89
+
90
+ logits_processors = make_logits_processors(repetition_penalty=1.2, repetition_context_size=256)
91
+ ```
92
+
93
+ Then pass `logits_processors=logits_processors` to `stream_generate()`.
94
+
95
+ ### Rule 4 — Use low temperature (0.3 – 0.5)
96
+
97
+ High temperature (> 0.7) produces incoherent, broken HTML. **Always use 0.3 – 0.5**.
98
+
99
+ ```python
100
+ from mlx_lm.sample_utils import make_sampler
101
+
102
+ sampler = make_sampler(temp=0.4) # ✅ Recommended
103
+ ```
104
+
105
+ ### Rule 5 — Post-process the output
106
+
107
+ The model may occasionally prepend training artifacts (system prompt) before the HTML. **Always clean the output:**
108
+
109
+ ```python
110
+ import re
111
+
112
+ def clean_html(text: str) -> str:
113
+ """Extract clean HTML from model output."""
114
+ # Remove leaked system prompts
115
+ text = re.sub(r"You are (?:Deep|Web[iI])coder.*?production-ready code\.\n*", "", text, flags=re.DOTALL)
116
+ text = re.sub(r"### Instruction:.*", "", text, flags=re.DOTALL)
117
+ text = re.sub(r"### Response:\s*", "", text, flags=re.DOTALL)
118
+
119
+ # Extract HTML document
120
+ match = re.search(r"(<(?:!DOCTYPE\s+html|html)[\s\S]*?</html>)", text, re.IGNORECASE)
121
+ if match:
122
+ return match.group(1).strip()
123
+
124
+ # Fallback
125
+ start = re.search(r"<(?:!DOCTYPE|html|head|body)", text, re.IGNORECASE)
126
+ if start:
127
+ html = text[start.start():].strip()
128
+ if not html.lower().startswith("<!doctype"):
129
+ html = "<!DOCTYPE html>\n<html>\n" + html + "\n</html>"
130
+ return html
131
+
132
+ return text.strip()
133
+ ```
134
+
135
+ ---
136
+
137
+ ## Quick Start — Complete Working Example
138
+
139
+ Copy-paste this and it will work:
140
+
141
+ ```python
142
+ from mlx_lm import load, stream_generate
143
+ from mlx_lm.sample_utils import make_sampler, make_logits_processors
144
+ import re
145
+
146
+ # 1. Load model
147
+ model, tokenizer = load("YOUR_USERNAME/WebICoder-v3-MLX-8bit")
148
+
149
+ # 2. Format prompt (MANDATORY)
150
+ user_prompt = "Create a modern portfolio website with a hero, project cards, and a contact form"
151
+
152
+ prompt = f"""### Instruction:
153
+ {user_prompt}
154
+
155
+ ### Response:
156
+ """
157
+
158
+ # 3. Configure sampler + repetition penalty (MANDATORY)
159
+ sampler = make_sampler(temp=0.4)
160
+ logits_processors = make_logits_processors(repetition_penalty=1.2, repetition_context_size=256)
161
+
162
+ # 4. Generate with stop at </html> (MANDATORY)
163
+ full_text = ""
164
+ for response in stream_generate(
165
+ model, tokenizer,
166
+ prompt=prompt,
167
+ max_tokens=4096,
168
+ sampler=sampler,
169
+ logits_processors=logits_processors,
170
+ ):
171
+ full_text += response.text
172
+ print(response.text, end="", flush=True)
173
+
174
+ if "</html>" in full_text or response.finish_reason:
175
+ break
176
+
177
+ # 5. Clean output (MANDATORY)
178
+ def clean_html(text):
179
+ text = re.sub(r"You are (?:Deep|Web[iI])coder.*?production-ready code\.\n*", "", text, flags=re.DOTALL)
180
+ match = re.search(r"(<(?:!DOCTYPE\s+html|html)[\s\S]*?</html>)", text, re.IGNORECASE)
181
+ return match.group(1).strip() if match else text.strip()
182
+
183
+ html = clean_html(full_text)
184
+
185
+ # Save to file
186
+ with open("output.html", "w") as f:
187
+ f.write(html)
188
+ print(f"\n\nSaved to output.html ({len(html)} chars)")
189
+ ```
190
+
191
+ ---
192
+
193
+ ## Recommended Parameters Summary
194
+
195
+ | Parameter | Value | Mandatory? |
196
+ |---|---|:---:|
197
+ | **Prompt format** | `### Instruction:` / `### Response:` | ✅ YES |
198
+ | **Temperature** | 0.3 – 0.5 | ✅ YES |
199
+ | **Repetition Penalty** | 1.2 | ✅ YES |
200
+ | **Repetition Context** | 256 | ✅ YES |
201
+ | **Max Tokens** | 4096 | ✅ YES |
202
+ | **Stop at `</html>`** | Check output and break | ✅ YES |
203
+ | **Post-processing** | `clean_html()` function | ✅ YES |
204
+ | **Top-p** | 0.9 | Recommended |
205
+ | **Top-k** | 50 | Optional |
206
+
207
+ ---
208
+
209
+ ## Using the Chat Template
210
+
211
+ The tokenizer includes a built-in chat template that handles prompt formatting automatically:
212
+
213
+ ```python
214
+ messages = [
215
+ {"role": "user", "content": "Create a dark-themed portfolio website with project cards"}
216
+ ]
217
+
218
+ prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
219
+ # This automatically wraps it in ### Instruction: / ### Response: format
220
+ ```
221
+
222
+ ## Using the Example Script
223
+
224
+ ```bash
225
+ # Single prompt
226
+ python example.py "Create a landing page for a coffee shop"
227
+
228
+ # Interactive mode
229
+ python example.py --interactive
230
+ ```
231
+
232
+ ---
233
+
234
+ ## Example Outputs
235
+
236
+ | Prompt | What You Get |
237
+ |---|---|
238
+ | "Create a portfolio with a hero and project cards" | Nav, animated hero, glassmorphism cards, contact form, footer |
239
+ | "Create a landing page for a fitness app" | Hero gradient, feature cards, testimonials, CTA, footer |
240
+ | "Create a pricing page with 3 tiers" | Toggle monthly/yearly, feature lists, highlighted plan |
241
+ | "Create a login page with split layout" | Gradient left, form right, social login buttons |
242
+
243
+ ---
244
+
245
+ ## What the Model Generates
246
+
247
+ When properly configured, WebICoder v3 produces:
248
+
249
+ - ✅ Complete `<!DOCTYPE html>` with `<head>`, `<meta>`, `<title>`
250
+ - ✅ **Vanilla CSS** — custom properties, gradients, glassmorphism, `backdrop-filter`
251
+ - ✅ **Responsive design** — `@media` queries, `clamp()`, CSS Grid `auto-fit`
252
+ - ✅ **Animations** — `fade-in` with `IntersectionObserver`, hover transitions
253
+ - ✅ **Modern design** — gradient text, blur effects, rounded corners, shadows
254
+ - ✅ **Complete pages** — nav, hero, content sections, footer
255
+
256
+ ---
257
+
258
+ ## Limitations
259
+
260
+ - Optimized for **single-page HTML** with embedded CSS/JS
261
+ - Context window: **4096 tokens** — very complex multi-section pages may still be truncated
262
+ - Based on Phi-2 (2.7B) — larger models will produce more sophisticated output
263
+ - English prompts work best
264
+
265
+ ---
266
+
267
+ ## Training Details
268
+
269
+ | Property | Value |
270
+ |---|---|
271
+ | **Base Model** | microsoft/phi-2 |
272
+ | **Fine-tuning** | Full fine-tuning on HTML/CSS code pairs |
273
+ | **Training Format** | Alpaca-style (Instruction / Response) |
274
+ | **Training Context** | 4096 tokens |
275
+ | **Precision** | float16 |
276
+ | **Quantization** | Post-training 8-bit (MLX affine, group_size=64) |
277
+
278
+ ---
279
+
280
+ ## Files Included
281
+
282
+ | File | Description |
283
+ |---|---|
284
+ | `model.safetensors` | Quantized model weights |
285
+ | `config.json` | Model architecture configuration |
286
+ | `tokenizer.json` | Tokenizer vocabulary |
287
+ | `tokenizer_config.json` | Tokenizer settings with chat template |
288
+ | `generation_config.json` | Recommended generation parameters |
289
+ | `example.py` | Ready-to-use example script with all mandatory rules |
290
+ | `LICENSE` | MIT License |
291
+
292
+ ---
293
+
294
+ ## Citation
295
+
296
+ ```bibtex
297
+ @misc{webicoder-v3,
298
+ title={WebICoder v3: Fine-tuned Phi-2 for HTML Code Generation},
299
+ year={2025},
300
+ publisher={Hugging Face},
301
+ url={https://huggingface.co/YOUR_USERNAME/WebICoder-v3-MLX-8bit}
302
+ }
303
+ ```
304
+
305
+ ## License
306
+
307
+ MIT License — see [LICENSE](LICENSE) for details.
added_tokens.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "\t\t": 50294,
3
+ "\t\t\t": 50293,
4
+ "\t\t\t\t": 50292,
5
+ "\t\t\t\t\t": 50291,
6
+ "\t\t\t\t\t\t": 50290,
7
+ "\t\t\t\t\t\t\t": 50289,
8
+ "\t\t\t\t\t\t\t\t": 50288,
9
+ "\t\t\t\t\t\t\t\t\t": 50287,
10
+ " ": 50286,
11
+ " ": 50285,
12
+ " ": 50284,
13
+ " ": 50283,
14
+ " ": 50282,
15
+ " ": 50281,
16
+ " ": 50280,
17
+ " ": 50279,
18
+ " ": 50278,
19
+ " ": 50277,
20
+ " ": 50276,
21
+ " ": 50275,
22
+ " ": 50274,
23
+ " ": 50273,
24
+ " ": 50272,
25
+ " ": 50271,
26
+ " ": 50270,
27
+ " ": 50269,
28
+ " ": 50268,
29
+ " ": 50267,
30
+ " ": 50266,
31
+ " ": 50265,
32
+ " ": 50264,
33
+ " ": 50263,
34
+ " ": 50262,
35
+ " ": 50261,
36
+ " ": 50260,
37
+ " ": 50259,
38
+ " ": 50258,
39
+ " ": 50257
40
+ }
config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "PhiForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 50256,
7
+ "dtype": "float16",
8
+ "embd_pdrop": 0.0,
9
+ "eos_token_id": 50256,
10
+ "hidden_act": "gelu_new",
11
+ "hidden_size": 2560,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 10240,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 4096,
16
+ "model_type": "phi",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 32,
19
+ "num_key_value_heads": 32,
20
+ "pad_token_id": 50256,
21
+ "partial_rotary_factor": 0.4,
22
+ "qk_layernorm": false,
23
+ "quantization": {
24
+ "group_size": 64,
25
+ "bits": 8,
26
+ "mode": "affine"
27
+ },
28
+ "quantization_config": {
29
+ "group_size": 64,
30
+ "bits": 8,
31
+ "mode": "affine"
32
+ },
33
+ "resid_pdrop": 0.1,
34
+ "rope_parameters": {
35
+ "partial_rotary_factor": 0.4,
36
+ "rope_theta": 10000.0,
37
+ "rope_type": "default"
38
+ },
39
+ "tie_word_embeddings": false,
40
+ "transformers_version": "5.0.0",
41
+ "use_cache": true,
42
+ "vocab_size": 51200
43
+ }
example.py ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """
3
+ WebICoder v3 — Quick Start Example
4
+ Generate HTML websites from natural language prompts using MLX on Apple Silicon.
5
+
6
+ ⚠️ MANDATORY: This script implements all 5 required rules for correct output.
7
+ See README.md for full documentation.
8
+
9
+ Usage:
10
+ python example.py "Create a landing page for a coffee shop"
11
+ python example.py --interactive
12
+ """
13
+
14
+ import sys
15
+ import re
16
+
17
+ from mlx_lm import load, stream_generate
18
+ from mlx_lm.sample_utils import make_sampler, make_logits_processors
19
+
20
+
21
+ # ─── Configuration ──────────────────────────────────────────────────────────
22
+ MODEL_PATH = "." # Current directory (the model repo)
23
+
24
+ # RULE 1: System prompt + Alpaca format (### Instruction / ### Response)
25
+ SYSTEM_PROMPT = (
26
+ "You are WebICoder, an expert frontend web developer specializing in premium, "
27
+ "Apple-inspired design. You create stunning websites using only HTML, CSS, and "
28
+ "vanilla JavaScript. Your designs feature: minimalist layouts, elegant typography, "
29
+ "smooth animations, glassmorphism effects, generous whitespace, and a refined "
30
+ "color palette. You always produce complete, production-ready code."
31
+ )
32
+
33
+ # RULE 2: Stop sequences — MANDATORY to prevent infinite loops
34
+ STOP_SEQUENCES = ["</html>", "### Instruction:", "You are Deepcoder", "You are WebICoder"]
35
+
36
+ # RULE 4: Low temperature — MANDATORY for coherent HTML
37
+ DEFAULT_TEMP = 0.4
38
+ DEFAULT_MAX_TOKENS = 4096
39
+
40
+
41
+ # ─── RULE 1: Prompt Formatting (MANDATORY) ──────────────────────────────────
42
+
43
+ def format_prompt(user_input: str) -> str:
44
+ """
45
+ MANDATORY: Format user input into the model's training prompt format.
46
+
47
+ The model was trained with Alpaca-style prompts. Sending raw text
48
+ without this formatting will produce garbage output.
49
+ """
50
+ return f"{SYSTEM_PROMPT}\n\n### Instruction:\n{user_input}\n\n### Response:\n"
51
+
52
+
53
+ # ─── RULE 5: Post-Processing (MANDATORY) ────────────────────────────────────
54
+
55
+ def clean_html(text: str) -> str:
56
+ """
57
+ MANDATORY: Extract clean HTML from model output.
58
+
59
+ The model may leak training artifacts (system prompt, instruction markers).
60
+ This function strips them and returns only valid HTML.
61
+ """
62
+ # Remove system prompt leaks
63
+ for pattern in [
64
+ r"You are (?:Deep|Web[iI])coder.*?production-ready code\.\n*",
65
+ r"### Instruction:.*",
66
+ r"### Response:\s*",
67
+ ]:
68
+ text = re.sub(pattern, "", text, flags=re.DOTALL)
69
+
70
+ # Extract complete HTML document
71
+ html_match = re.search(r"(<(?:!DOCTYPE\s+html|html)[\s\S]*?</html>)", text, re.IGNORECASE)
72
+ if html_match:
73
+ return html_match.group(1).strip()
74
+
75
+ # Fallback: find any HTML content and wrap it
76
+ html_start = re.search(r"<(?:!DOCTYPE|html|head|body|link)", text, re.IGNORECASE)
77
+ if html_start:
78
+ html = text[html_start.start():].strip()
79
+ if not html.lower().startswith("<!doctype"):
80
+ html = "<!DOCTYPE html>\n<html>\n" + html
81
+ if "</html>" not in html.lower():
82
+ html += "\n</html>"
83
+ return html
84
+
85
+ return text.strip()
86
+
87
+
88
+ # ─── Generation ─────────────────────────────────────────────────────────────
89
+
90
+ def generate_html(prompt: str, temperature: float = DEFAULT_TEMP, max_tokens: int = DEFAULT_MAX_TOKENS) -> str:
91
+ """
92
+ Generate HTML from a natural language prompt.
93
+
94
+ Implements all 5 mandatory rules:
95
+ 1. Prompt formatting (### Instruction / ### Response)
96
+ 2. Stop at </html>
97
+ 3. Repetition penalty (1.2, context=256)
98
+ 4. Low temperature (0.4)
99
+ 5. Post-processing (clean_html)
100
+ """
101
+ print(f"[INFO] Loading model from: {MODEL_PATH}")
102
+ model, tokenizer = load(MODEL_PATH)
103
+
104
+ # RULE 1: Format the prompt
105
+ formatted_prompt = format_prompt(prompt)
106
+
107
+ # RULE 4: Low temperature sampler
108
+ sampler = make_sampler(temp=temperature)
109
+
110
+ # RULE 3: Repetition penalty — MANDATORY
111
+ logits_processors = make_logits_processors(
112
+ repetition_penalty=1.2,
113
+ repetition_context_size=256,
114
+ )
115
+
116
+ print(f"[INFO] Generating (temp={temperature}, max_tokens={max_tokens}, rep_penalty=1.2)...")
117
+ print("─" * 60)
118
+
119
+ full_text = ""
120
+ last_response = None
121
+
122
+ for response in stream_generate(
123
+ model, tokenizer,
124
+ prompt=formatted_prompt,
125
+ max_tokens=max_tokens,
126
+ sampler=sampler,
127
+ logits_processors=logits_processors, # RULE 3
128
+ ):
129
+ last_response = response
130
+ token_str = response.text
131
+ full_text += token_str
132
+ print(token_str, end="", flush=True)
133
+
134
+ # RULE 2: Stop at </html> — MANDATORY
135
+ should_stop = False
136
+ for stop_seq in STOP_SEQUENCES:
137
+ if stop_seq in full_text:
138
+ idx = full_text.find(stop_seq)
139
+ if stop_seq == "</html>":
140
+ full_text = full_text[:idx + len(stop_seq)]
141
+ else:
142
+ full_text = full_text[:idx]
143
+ should_stop = True
144
+ break
145
+
146
+ if should_stop or response.finish_reason is not None:
147
+ break
148
+
149
+ print("\n" + "─" * 60)
150
+ if last_response:
151
+ print(f"[INFO] Generated {last_response.generation_tokens} tokens at {last_response.generation_tps:.1f} tok/s")
152
+ print(f"[INFO] Peak memory: {last_response.peak_memory:.2f} GB")
153
+
154
+ # RULE 5: Clean the output — MANDATORY
155
+ return clean_html(full_text)
156
+
157
+
158
+ # ─── Main ────────────────────────────────────────────────────────────────────
159
+
160
+ def main():
161
+ if len(sys.argv) > 1 and sys.argv[1] != "--interactive":
162
+ # Single prompt mode
163
+ prompt = " ".join(sys.argv[1:])
164
+ html = generate_html(prompt)
165
+
166
+ output_file = "output.html"
167
+ with open(output_file, "w") as f:
168
+ f.write(html)
169
+ print(f"\n[INFO] Saved to {output_file} ({len(html)} chars)")
170
+
171
+ else:
172
+ # Interactive mode
173
+ print("=" * 60)
174
+ print(" ⚡ WebICoder v3 — Interactive Mode")
175
+ print(" Type a website description, press Enter to generate.")
176
+ print(" Type 'quit' to exit.")
177
+ print("=" * 60)
178
+
179
+ while True:
180
+ try:
181
+ prompt = input("\n🌐 Describe your website: ").strip()
182
+ if not prompt or prompt.lower() in ("quit", "exit", "q"):
183
+ break
184
+
185
+ html = generate_html(prompt)
186
+
187
+ output_file = "output.html"
188
+ with open(output_file, "w") as f:
189
+ f.write(html)
190
+ print(f"\n[INFO] Saved to {output_file} ({len(html)} chars)")
191
+
192
+ except KeyboardInterrupt:
193
+ print("\n[INFO] Bye!")
194
+ break
195
+
196
+
197
+ if __name__ == "__main__":
198
+ main()
generation_config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "do_sample": true,
6
+ "temperature": 0.4,
7
+ "top_p": 0.9,
8
+ "top_k": 50,
9
+ "repetition_penalty": 1.15,
10
+ "max_new_tokens": 4096,
11
+ "stop_strings": [
12
+ "</html>",
13
+ "### Instruction:",
14
+ "You are Deepcoder",
15
+ "You are WebICoder"
16
+ ],
17
+ "transformers_version": "5.0.0"
18
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4224a56beffcbf57ca9eebf849ae793dfae5f931d778dc9ff97b50e77639f00c
3
+ size 2954404123
model.safetensors.index.json ADDED
@@ -0,0 +1,849 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 2954311680,
4
+ "total_parameters": 2779683840
5
+ },
6
+ "weight_map": {
7
+ "lm_head.bias": "model.safetensors",
8
+ "lm_head.biases": "model.safetensors",
9
+ "lm_head.scales": "model.safetensors",
10
+ "lm_head.weight": "model.safetensors",
11
+ "model.embed_tokens.biases": "model.safetensors",
12
+ "model.embed_tokens.scales": "model.safetensors",
13
+ "model.embed_tokens.weight": "model.safetensors",
14
+ "model.final_layernorm.bias": "model.safetensors",
15
+ "model.final_layernorm.weight": "model.safetensors",
16
+ "model.layers.0.input_layernorm.bias": "model.safetensors",
17
+ "model.layers.0.input_layernorm.weight": "model.safetensors",
18
+ "model.layers.0.mlp.fc1.bias": "model.safetensors",
19
+ "model.layers.0.mlp.fc1.biases": "model.safetensors",
20
+ "model.layers.0.mlp.fc1.scales": "model.safetensors",
21
+ "model.layers.0.mlp.fc1.weight": "model.safetensors",
22
+ "model.layers.0.mlp.fc2.bias": "model.safetensors",
23
+ "model.layers.0.mlp.fc2.biases": "model.safetensors",
24
+ "model.layers.0.mlp.fc2.scales": "model.safetensors",
25
+ "model.layers.0.mlp.fc2.weight": "model.safetensors",
26
+ "model.layers.0.self_attn.dense.bias": "model.safetensors",
27
+ "model.layers.0.self_attn.dense.biases": "model.safetensors",
28
+ "model.layers.0.self_attn.dense.scales": "model.safetensors",
29
+ "model.layers.0.self_attn.dense.weight": "model.safetensors",
30
+ "model.layers.0.self_attn.k_proj.bias": "model.safetensors",
31
+ "model.layers.0.self_attn.k_proj.biases": "model.safetensors",
32
+ "model.layers.0.self_attn.k_proj.scales": "model.safetensors",
33
+ "model.layers.0.self_attn.k_proj.weight": "model.safetensors",
34
+ "model.layers.0.self_attn.q_proj.bias": "model.safetensors",
35
+ "model.layers.0.self_attn.q_proj.biases": "model.safetensors",
36
+ "model.layers.0.self_attn.q_proj.scales": "model.safetensors",
37
+ "model.layers.0.self_attn.q_proj.weight": "model.safetensors",
38
+ "model.layers.0.self_attn.v_proj.bias": "model.safetensors",
39
+ "model.layers.0.self_attn.v_proj.biases": "model.safetensors",
40
+ "model.layers.0.self_attn.v_proj.scales": "model.safetensors",
41
+ "model.layers.0.self_attn.v_proj.weight": "model.safetensors",
42
+ "model.layers.1.input_layernorm.bias": "model.safetensors",
43
+ "model.layers.1.input_layernorm.weight": "model.safetensors",
44
+ "model.layers.1.mlp.fc1.bias": "model.safetensors",
45
+ "model.layers.1.mlp.fc1.biases": "model.safetensors",
46
+ "model.layers.1.mlp.fc1.scales": "model.safetensors",
47
+ "model.layers.1.mlp.fc1.weight": "model.safetensors",
48
+ "model.layers.1.mlp.fc2.bias": "model.safetensors",
49
+ "model.layers.1.mlp.fc2.biases": "model.safetensors",
50
+ "model.layers.1.mlp.fc2.scales": "model.safetensors",
51
+ "model.layers.1.mlp.fc2.weight": "model.safetensors",
52
+ "model.layers.1.self_attn.dense.bias": "model.safetensors",
53
+ "model.layers.1.self_attn.dense.biases": "model.safetensors",
54
+ "model.layers.1.self_attn.dense.scales": "model.safetensors",
55
+ "model.layers.1.self_attn.dense.weight": "model.safetensors",
56
+ "model.layers.1.self_attn.k_proj.bias": "model.safetensors",
57
+ "model.layers.1.self_attn.k_proj.biases": "model.safetensors",
58
+ "model.layers.1.self_attn.k_proj.scales": "model.safetensors",
59
+ "model.layers.1.self_attn.k_proj.weight": "model.safetensors",
60
+ "model.layers.1.self_attn.q_proj.bias": "model.safetensors",
61
+ "model.layers.1.self_attn.q_proj.biases": "model.safetensors",
62
+ "model.layers.1.self_attn.q_proj.scales": "model.safetensors",
63
+ "model.layers.1.self_attn.q_proj.weight": "model.safetensors",
64
+ "model.layers.1.self_attn.v_proj.bias": "model.safetensors",
65
+ "model.layers.1.self_attn.v_proj.biases": "model.safetensors",
66
+ "model.layers.1.self_attn.v_proj.scales": "model.safetensors",
67
+ "model.layers.1.self_attn.v_proj.weight": "model.safetensors",
68
+ "model.layers.10.input_layernorm.bias": "model.safetensors",
69
+ "model.layers.10.input_layernorm.weight": "model.safetensors",
70
+ "model.layers.10.mlp.fc1.bias": "model.safetensors",
71
+ "model.layers.10.mlp.fc1.biases": "model.safetensors",
72
+ "model.layers.10.mlp.fc1.scales": "model.safetensors",
73
+ "model.layers.10.mlp.fc1.weight": "model.safetensors",
74
+ "model.layers.10.mlp.fc2.bias": "model.safetensors",
75
+ "model.layers.10.mlp.fc2.biases": "model.safetensors",
76
+ "model.layers.10.mlp.fc2.scales": "model.safetensors",
77
+ "model.layers.10.mlp.fc2.weight": "model.safetensors",
78
+ "model.layers.10.self_attn.dense.bias": "model.safetensors",
79
+ "model.layers.10.self_attn.dense.biases": "model.safetensors",
80
+ "model.layers.10.self_attn.dense.scales": "model.safetensors",
81
+ "model.layers.10.self_attn.dense.weight": "model.safetensors",
82
+ "model.layers.10.self_attn.k_proj.bias": "model.safetensors",
83
+ "model.layers.10.self_attn.k_proj.biases": "model.safetensors",
84
+ "model.layers.10.self_attn.k_proj.scales": "model.safetensors",
85
+ "model.layers.10.self_attn.k_proj.weight": "model.safetensors",
86
+ "model.layers.10.self_attn.q_proj.bias": "model.safetensors",
87
+ "model.layers.10.self_attn.q_proj.biases": "model.safetensors",
88
+ "model.layers.10.self_attn.q_proj.scales": "model.safetensors",
89
+ "model.layers.10.self_attn.q_proj.weight": "model.safetensors",
90
+ "model.layers.10.self_attn.v_proj.bias": "model.safetensors",
91
+ "model.layers.10.self_attn.v_proj.biases": "model.safetensors",
92
+ "model.layers.10.self_attn.v_proj.scales": "model.safetensors",
93
+ "model.layers.10.self_attn.v_proj.weight": "model.safetensors",
94
+ "model.layers.11.input_layernorm.bias": "model.safetensors",
95
+ "model.layers.11.input_layernorm.weight": "model.safetensors",
96
+ "model.layers.11.mlp.fc1.bias": "model.safetensors",
97
+ "model.layers.11.mlp.fc1.biases": "model.safetensors",
98
+ "model.layers.11.mlp.fc1.scales": "model.safetensors",
99
+ "model.layers.11.mlp.fc1.weight": "model.safetensors",
100
+ "model.layers.11.mlp.fc2.bias": "model.safetensors",
101
+ "model.layers.11.mlp.fc2.biases": "model.safetensors",
102
+ "model.layers.11.mlp.fc2.scales": "model.safetensors",
103
+ "model.layers.11.mlp.fc2.weight": "model.safetensors",
104
+ "model.layers.11.self_attn.dense.bias": "model.safetensors",
105
+ "model.layers.11.self_attn.dense.biases": "model.safetensors",
106
+ "model.layers.11.self_attn.dense.scales": "model.safetensors",
107
+ "model.layers.11.self_attn.dense.weight": "model.safetensors",
108
+ "model.layers.11.self_attn.k_proj.bias": "model.safetensors",
109
+ "model.layers.11.self_attn.k_proj.biases": "model.safetensors",
110
+ "model.layers.11.self_attn.k_proj.scales": "model.safetensors",
111
+ "model.layers.11.self_attn.k_proj.weight": "model.safetensors",
112
+ "model.layers.11.self_attn.q_proj.bias": "model.safetensors",
113
+ "model.layers.11.self_attn.q_proj.biases": "model.safetensors",
114
+ "model.layers.11.self_attn.q_proj.scales": "model.safetensors",
115
+ "model.layers.11.self_attn.q_proj.weight": "model.safetensors",
116
+ "model.layers.11.self_attn.v_proj.bias": "model.safetensors",
117
+ "model.layers.11.self_attn.v_proj.biases": "model.safetensors",
118
+ "model.layers.11.self_attn.v_proj.scales": "model.safetensors",
119
+ "model.layers.11.self_attn.v_proj.weight": "model.safetensors",
120
+ "model.layers.12.input_layernorm.bias": "model.safetensors",
121
+ "model.layers.12.input_layernorm.weight": "model.safetensors",
122
+ "model.layers.12.mlp.fc1.bias": "model.safetensors",
123
+ "model.layers.12.mlp.fc1.biases": "model.safetensors",
124
+ "model.layers.12.mlp.fc1.scales": "model.safetensors",
125
+ "model.layers.12.mlp.fc1.weight": "model.safetensors",
126
+ "model.layers.12.mlp.fc2.bias": "model.safetensors",
127
+ "model.layers.12.mlp.fc2.biases": "model.safetensors",
128
+ "model.layers.12.mlp.fc2.scales": "model.safetensors",
129
+ "model.layers.12.mlp.fc2.weight": "model.safetensors",
130
+ "model.layers.12.self_attn.dense.bias": "model.safetensors",
131
+ "model.layers.12.self_attn.dense.biases": "model.safetensors",
132
+ "model.layers.12.self_attn.dense.scales": "model.safetensors",
133
+ "model.layers.12.self_attn.dense.weight": "model.safetensors",
134
+ "model.layers.12.self_attn.k_proj.bias": "model.safetensors",
135
+ "model.layers.12.self_attn.k_proj.biases": "model.safetensors",
136
+ "model.layers.12.self_attn.k_proj.scales": "model.safetensors",
137
+ "model.layers.12.self_attn.k_proj.weight": "model.safetensors",
138
+ "model.layers.12.self_attn.q_proj.bias": "model.safetensors",
139
+ "model.layers.12.self_attn.q_proj.biases": "model.safetensors",
140
+ "model.layers.12.self_attn.q_proj.scales": "model.safetensors",
141
+ "model.layers.12.self_attn.q_proj.weight": "model.safetensors",
142
+ "model.layers.12.self_attn.v_proj.bias": "model.safetensors",
143
+ "model.layers.12.self_attn.v_proj.biases": "model.safetensors",
144
+ "model.layers.12.self_attn.v_proj.scales": "model.safetensors",
145
+ "model.layers.12.self_attn.v_proj.weight": "model.safetensors",
146
+ "model.layers.13.input_layernorm.bias": "model.safetensors",
147
+ "model.layers.13.input_layernorm.weight": "model.safetensors",
148
+ "model.layers.13.mlp.fc1.bias": "model.safetensors",
149
+ "model.layers.13.mlp.fc1.biases": "model.safetensors",
150
+ "model.layers.13.mlp.fc1.scales": "model.safetensors",
151
+ "model.layers.13.mlp.fc1.weight": "model.safetensors",
152
+ "model.layers.13.mlp.fc2.bias": "model.safetensors",
153
+ "model.layers.13.mlp.fc2.biases": "model.safetensors",
154
+ "model.layers.13.mlp.fc2.scales": "model.safetensors",
155
+ "model.layers.13.mlp.fc2.weight": "model.safetensors",
156
+ "model.layers.13.self_attn.dense.bias": "model.safetensors",
157
+ "model.layers.13.self_attn.dense.biases": "model.safetensors",
158
+ "model.layers.13.self_attn.dense.scales": "model.safetensors",
159
+ "model.layers.13.self_attn.dense.weight": "model.safetensors",
160
+ "model.layers.13.self_attn.k_proj.bias": "model.safetensors",
161
+ "model.layers.13.self_attn.k_proj.biases": "model.safetensors",
162
+ "model.layers.13.self_attn.k_proj.scales": "model.safetensors",
163
+ "model.layers.13.self_attn.k_proj.weight": "model.safetensors",
164
+ "model.layers.13.self_attn.q_proj.bias": "model.safetensors",
165
+ "model.layers.13.self_attn.q_proj.biases": "model.safetensors",
166
+ "model.layers.13.self_attn.q_proj.scales": "model.safetensors",
167
+ "model.layers.13.self_attn.q_proj.weight": "model.safetensors",
168
+ "model.layers.13.self_attn.v_proj.bias": "model.safetensors",
169
+ "model.layers.13.self_attn.v_proj.biases": "model.safetensors",
170
+ "model.layers.13.self_attn.v_proj.scales": "model.safetensors",
171
+ "model.layers.13.self_attn.v_proj.weight": "model.safetensors",
172
+ "model.layers.14.input_layernorm.bias": "model.safetensors",
173
+ "model.layers.14.input_layernorm.weight": "model.safetensors",
174
+ "model.layers.14.mlp.fc1.bias": "model.safetensors",
175
+ "model.layers.14.mlp.fc1.biases": "model.safetensors",
176
+ "model.layers.14.mlp.fc1.scales": "model.safetensors",
177
+ "model.layers.14.mlp.fc1.weight": "model.safetensors",
178
+ "model.layers.14.mlp.fc2.bias": "model.safetensors",
179
+ "model.layers.14.mlp.fc2.biases": "model.safetensors",
180
+ "model.layers.14.mlp.fc2.scales": "model.safetensors",
181
+ "model.layers.14.mlp.fc2.weight": "model.safetensors",
182
+ "model.layers.14.self_attn.dense.bias": "model.safetensors",
183
+ "model.layers.14.self_attn.dense.biases": "model.safetensors",
184
+ "model.layers.14.self_attn.dense.scales": "model.safetensors",
185
+ "model.layers.14.self_attn.dense.weight": "model.safetensors",
186
+ "model.layers.14.self_attn.k_proj.bias": "model.safetensors",
187
+ "model.layers.14.self_attn.k_proj.biases": "model.safetensors",
188
+ "model.layers.14.self_attn.k_proj.scales": "model.safetensors",
189
+ "model.layers.14.self_attn.k_proj.weight": "model.safetensors",
190
+ "model.layers.14.self_attn.q_proj.bias": "model.safetensors",
191
+ "model.layers.14.self_attn.q_proj.biases": "model.safetensors",
192
+ "model.layers.14.self_attn.q_proj.scales": "model.safetensors",
193
+ "model.layers.14.self_attn.q_proj.weight": "model.safetensors",
194
+ "model.layers.14.self_attn.v_proj.bias": "model.safetensors",
195
+ "model.layers.14.self_attn.v_proj.biases": "model.safetensors",
196
+ "model.layers.14.self_attn.v_proj.scales": "model.safetensors",
197
+ "model.layers.14.self_attn.v_proj.weight": "model.safetensors",
198
+ "model.layers.15.input_layernorm.bias": "model.safetensors",
199
+ "model.layers.15.input_layernorm.weight": "model.safetensors",
200
+ "model.layers.15.mlp.fc1.bias": "model.safetensors",
201
+ "model.layers.15.mlp.fc1.biases": "model.safetensors",
202
+ "model.layers.15.mlp.fc1.scales": "model.safetensors",
203
+ "model.layers.15.mlp.fc1.weight": "model.safetensors",
204
+ "model.layers.15.mlp.fc2.bias": "model.safetensors",
205
+ "model.layers.15.mlp.fc2.biases": "model.safetensors",
206
+ "model.layers.15.mlp.fc2.scales": "model.safetensors",
207
+ "model.layers.15.mlp.fc2.weight": "model.safetensors",
208
+ "model.layers.15.self_attn.dense.bias": "model.safetensors",
209
+ "model.layers.15.self_attn.dense.biases": "model.safetensors",
210
+ "model.layers.15.self_attn.dense.scales": "model.safetensors",
211
+ "model.layers.15.self_attn.dense.weight": "model.safetensors",
212
+ "model.layers.15.self_attn.k_proj.bias": "model.safetensors",
213
+ "model.layers.15.self_attn.k_proj.biases": "model.safetensors",
214
+ "model.layers.15.self_attn.k_proj.scales": "model.safetensors",
215
+ "model.layers.15.self_attn.k_proj.weight": "model.safetensors",
216
+ "model.layers.15.self_attn.q_proj.bias": "model.safetensors",
217
+ "model.layers.15.self_attn.q_proj.biases": "model.safetensors",
218
+ "model.layers.15.self_attn.q_proj.scales": "model.safetensors",
219
+ "model.layers.15.self_attn.q_proj.weight": "model.safetensors",
220
+ "model.layers.15.self_attn.v_proj.bias": "model.safetensors",
221
+ "model.layers.15.self_attn.v_proj.biases": "model.safetensors",
222
+ "model.layers.15.self_attn.v_proj.scales": "model.safetensors",
223
+ "model.layers.15.self_attn.v_proj.weight": "model.safetensors",
224
+ "model.layers.16.input_layernorm.bias": "model.safetensors",
225
+ "model.layers.16.input_layernorm.weight": "model.safetensors",
226
+ "model.layers.16.mlp.fc1.bias": "model.safetensors",
227
+ "model.layers.16.mlp.fc1.biases": "model.safetensors",
228
+ "model.layers.16.mlp.fc1.scales": "model.safetensors",
229
+ "model.layers.16.mlp.fc1.weight": "model.safetensors",
230
+ "model.layers.16.mlp.fc2.bias": "model.safetensors",
231
+ "model.layers.16.mlp.fc2.biases": "model.safetensors",
232
+ "model.layers.16.mlp.fc2.scales": "model.safetensors",
233
+ "model.layers.16.mlp.fc2.weight": "model.safetensors",
234
+ "model.layers.16.self_attn.dense.bias": "model.safetensors",
235
+ "model.layers.16.self_attn.dense.biases": "model.safetensors",
236
+ "model.layers.16.self_attn.dense.scales": "model.safetensors",
237
+ "model.layers.16.self_attn.dense.weight": "model.safetensors",
238
+ "model.layers.16.self_attn.k_proj.bias": "model.safetensors",
239
+ "model.layers.16.self_attn.k_proj.biases": "model.safetensors",
240
+ "model.layers.16.self_attn.k_proj.scales": "model.safetensors",
241
+ "model.layers.16.self_attn.k_proj.weight": "model.safetensors",
242
+ "model.layers.16.self_attn.q_proj.bias": "model.safetensors",
243
+ "model.layers.16.self_attn.q_proj.biases": "model.safetensors",
244
+ "model.layers.16.self_attn.q_proj.scales": "model.safetensors",
245
+ "model.layers.16.self_attn.q_proj.weight": "model.safetensors",
246
+ "model.layers.16.self_attn.v_proj.bias": "model.safetensors",
247
+ "model.layers.16.self_attn.v_proj.biases": "model.safetensors",
248
+ "model.layers.16.self_attn.v_proj.scales": "model.safetensors",
249
+ "model.layers.16.self_attn.v_proj.weight": "model.safetensors",
250
+ "model.layers.17.input_layernorm.bias": "model.safetensors",
251
+ "model.layers.17.input_layernorm.weight": "model.safetensors",
252
+ "model.layers.17.mlp.fc1.bias": "model.safetensors",
253
+ "model.layers.17.mlp.fc1.biases": "model.safetensors",
254
+ "model.layers.17.mlp.fc1.scales": "model.safetensors",
255
+ "model.layers.17.mlp.fc1.weight": "model.safetensors",
256
+ "model.layers.17.mlp.fc2.bias": "model.safetensors",
257
+ "model.layers.17.mlp.fc2.biases": "model.safetensors",
258
+ "model.layers.17.mlp.fc2.scales": "model.safetensors",
259
+ "model.layers.17.mlp.fc2.weight": "model.safetensors",
260
+ "model.layers.17.self_attn.dense.bias": "model.safetensors",
261
+ "model.layers.17.self_attn.dense.biases": "model.safetensors",
262
+ "model.layers.17.self_attn.dense.scales": "model.safetensors",
263
+ "model.layers.17.self_attn.dense.weight": "model.safetensors",
264
+ "model.layers.17.self_attn.k_proj.bias": "model.safetensors",
265
+ "model.layers.17.self_attn.k_proj.biases": "model.safetensors",
266
+ "model.layers.17.self_attn.k_proj.scales": "model.safetensors",
267
+ "model.layers.17.self_attn.k_proj.weight": "model.safetensors",
268
+ "model.layers.17.self_attn.q_proj.bias": "model.safetensors",
269
+ "model.layers.17.self_attn.q_proj.biases": "model.safetensors",
270
+ "model.layers.17.self_attn.q_proj.scales": "model.safetensors",
271
+ "model.layers.17.self_attn.q_proj.weight": "model.safetensors",
272
+ "model.layers.17.self_attn.v_proj.bias": "model.safetensors",
273
+ "model.layers.17.self_attn.v_proj.biases": "model.safetensors",
274
+ "model.layers.17.self_attn.v_proj.scales": "model.safetensors",
275
+ "model.layers.17.self_attn.v_proj.weight": "model.safetensors",
276
+ "model.layers.18.input_layernorm.bias": "model.safetensors",
277
+ "model.layers.18.input_layernorm.weight": "model.safetensors",
278
+ "model.layers.18.mlp.fc1.bias": "model.safetensors",
279
+ "model.layers.18.mlp.fc1.biases": "model.safetensors",
280
+ "model.layers.18.mlp.fc1.scales": "model.safetensors",
281
+ "model.layers.18.mlp.fc1.weight": "model.safetensors",
282
+ "model.layers.18.mlp.fc2.bias": "model.safetensors",
283
+ "model.layers.18.mlp.fc2.biases": "model.safetensors",
284
+ "model.layers.18.mlp.fc2.scales": "model.safetensors",
285
+ "model.layers.18.mlp.fc2.weight": "model.safetensors",
286
+ "model.layers.18.self_attn.dense.bias": "model.safetensors",
287
+ "model.layers.18.self_attn.dense.biases": "model.safetensors",
288
+ "model.layers.18.self_attn.dense.scales": "model.safetensors",
289
+ "model.layers.18.self_attn.dense.weight": "model.safetensors",
290
+ "model.layers.18.self_attn.k_proj.bias": "model.safetensors",
291
+ "model.layers.18.self_attn.k_proj.biases": "model.safetensors",
292
+ "model.layers.18.self_attn.k_proj.scales": "model.safetensors",
293
+ "model.layers.18.self_attn.k_proj.weight": "model.safetensors",
294
+ "model.layers.18.self_attn.q_proj.bias": "model.safetensors",
295
+ "model.layers.18.self_attn.q_proj.biases": "model.safetensors",
296
+ "model.layers.18.self_attn.q_proj.scales": "model.safetensors",
297
+ "model.layers.18.self_attn.q_proj.weight": "model.safetensors",
298
+ "model.layers.18.self_attn.v_proj.bias": "model.safetensors",
299
+ "model.layers.18.self_attn.v_proj.biases": "model.safetensors",
300
+ "model.layers.18.self_attn.v_proj.scales": "model.safetensors",
301
+ "model.layers.18.self_attn.v_proj.weight": "model.safetensors",
302
+ "model.layers.19.input_layernorm.bias": "model.safetensors",
303
+ "model.layers.19.input_layernorm.weight": "model.safetensors",
304
+ "model.layers.19.mlp.fc1.bias": "model.safetensors",
305
+ "model.layers.19.mlp.fc1.biases": "model.safetensors",
306
+ "model.layers.19.mlp.fc1.scales": "model.safetensors",
307
+ "model.layers.19.mlp.fc1.weight": "model.safetensors",
308
+ "model.layers.19.mlp.fc2.bias": "model.safetensors",
309
+ "model.layers.19.mlp.fc2.biases": "model.safetensors",
310
+ "model.layers.19.mlp.fc2.scales": "model.safetensors",
311
+ "model.layers.19.mlp.fc2.weight": "model.safetensors",
312
+ "model.layers.19.self_attn.dense.bias": "model.safetensors",
313
+ "model.layers.19.self_attn.dense.biases": "model.safetensors",
314
+ "model.layers.19.self_attn.dense.scales": "model.safetensors",
315
+ "model.layers.19.self_attn.dense.weight": "model.safetensors",
316
+ "model.layers.19.self_attn.k_proj.bias": "model.safetensors",
317
+ "model.layers.19.self_attn.k_proj.biases": "model.safetensors",
318
+ "model.layers.19.self_attn.k_proj.scales": "model.safetensors",
319
+ "model.layers.19.self_attn.k_proj.weight": "model.safetensors",
320
+ "model.layers.19.self_attn.q_proj.bias": "model.safetensors",
321
+ "model.layers.19.self_attn.q_proj.biases": "model.safetensors",
322
+ "model.layers.19.self_attn.q_proj.scales": "model.safetensors",
323
+ "model.layers.19.self_attn.q_proj.weight": "model.safetensors",
324
+ "model.layers.19.self_attn.v_proj.bias": "model.safetensors",
325
+ "model.layers.19.self_attn.v_proj.biases": "model.safetensors",
326
+ "model.layers.19.self_attn.v_proj.scales": "model.safetensors",
327
+ "model.layers.19.self_attn.v_proj.weight": "model.safetensors",
328
+ "model.layers.2.input_layernorm.bias": "model.safetensors",
329
+ "model.layers.2.input_layernorm.weight": "model.safetensors",
330
+ "model.layers.2.mlp.fc1.bias": "model.safetensors",
331
+ "model.layers.2.mlp.fc1.biases": "model.safetensors",
332
+ "model.layers.2.mlp.fc1.scales": "model.safetensors",
333
+ "model.layers.2.mlp.fc1.weight": "model.safetensors",
334
+ "model.layers.2.mlp.fc2.bias": "model.safetensors",
335
+ "model.layers.2.mlp.fc2.biases": "model.safetensors",
336
+ "model.layers.2.mlp.fc2.scales": "model.safetensors",
337
+ "model.layers.2.mlp.fc2.weight": "model.safetensors",
338
+ "model.layers.2.self_attn.dense.bias": "model.safetensors",
339
+ "model.layers.2.self_attn.dense.biases": "model.safetensors",
340
+ "model.layers.2.self_attn.dense.scales": "model.safetensors",
341
+ "model.layers.2.self_attn.dense.weight": "model.safetensors",
342
+ "model.layers.2.self_attn.k_proj.bias": "model.safetensors",
343
+ "model.layers.2.self_attn.k_proj.biases": "model.safetensors",
344
+ "model.layers.2.self_attn.k_proj.scales": "model.safetensors",
345
+ "model.layers.2.self_attn.k_proj.weight": "model.safetensors",
346
+ "model.layers.2.self_attn.q_proj.bias": "model.safetensors",
347
+ "model.layers.2.self_attn.q_proj.biases": "model.safetensors",
348
+ "model.layers.2.self_attn.q_proj.scales": "model.safetensors",
349
+ "model.layers.2.self_attn.q_proj.weight": "model.safetensors",
350
+ "model.layers.2.self_attn.v_proj.bias": "model.safetensors",
351
+ "model.layers.2.self_attn.v_proj.biases": "model.safetensors",
352
+ "model.layers.2.self_attn.v_proj.scales": "model.safetensors",
353
+ "model.layers.2.self_attn.v_proj.weight": "model.safetensors",
354
+ "model.layers.20.input_layernorm.bias": "model.safetensors",
355
+ "model.layers.20.input_layernorm.weight": "model.safetensors",
356
+ "model.layers.20.mlp.fc1.bias": "model.safetensors",
357
+ "model.layers.20.mlp.fc1.biases": "model.safetensors",
358
+ "model.layers.20.mlp.fc1.scales": "model.safetensors",
359
+ "model.layers.20.mlp.fc1.weight": "model.safetensors",
360
+ "model.layers.20.mlp.fc2.bias": "model.safetensors",
361
+ "model.layers.20.mlp.fc2.biases": "model.safetensors",
362
+ "model.layers.20.mlp.fc2.scales": "model.safetensors",
363
+ "model.layers.20.mlp.fc2.weight": "model.safetensors",
364
+ "model.layers.20.self_attn.dense.bias": "model.safetensors",
365
+ "model.layers.20.self_attn.dense.biases": "model.safetensors",
366
+ "model.layers.20.self_attn.dense.scales": "model.safetensors",
367
+ "model.layers.20.self_attn.dense.weight": "model.safetensors",
368
+ "model.layers.20.self_attn.k_proj.bias": "model.safetensors",
369
+ "model.layers.20.self_attn.k_proj.biases": "model.safetensors",
370
+ "model.layers.20.self_attn.k_proj.scales": "model.safetensors",
371
+ "model.layers.20.self_attn.k_proj.weight": "model.safetensors",
372
+ "model.layers.20.self_attn.q_proj.bias": "model.safetensors",
373
+ "model.layers.20.self_attn.q_proj.biases": "model.safetensors",
374
+ "model.layers.20.self_attn.q_proj.scales": "model.safetensors",
375
+ "model.layers.20.self_attn.q_proj.weight": "model.safetensors",
376
+ "model.layers.20.self_attn.v_proj.bias": "model.safetensors",
377
+ "model.layers.20.self_attn.v_proj.biases": "model.safetensors",
378
+ "model.layers.20.self_attn.v_proj.scales": "model.safetensors",
379
+ "model.layers.20.self_attn.v_proj.weight": "model.safetensors",
380
+ "model.layers.21.input_layernorm.bias": "model.safetensors",
381
+ "model.layers.21.input_layernorm.weight": "model.safetensors",
382
+ "model.layers.21.mlp.fc1.bias": "model.safetensors",
383
+ "model.layers.21.mlp.fc1.biases": "model.safetensors",
384
+ "model.layers.21.mlp.fc1.scales": "model.safetensors",
385
+ "model.layers.21.mlp.fc1.weight": "model.safetensors",
386
+ "model.layers.21.mlp.fc2.bias": "model.safetensors",
387
+ "model.layers.21.mlp.fc2.biases": "model.safetensors",
388
+ "model.layers.21.mlp.fc2.scales": "model.safetensors",
389
+ "model.layers.21.mlp.fc2.weight": "model.safetensors",
390
+ "model.layers.21.self_attn.dense.bias": "model.safetensors",
391
+ "model.layers.21.self_attn.dense.biases": "model.safetensors",
392
+ "model.layers.21.self_attn.dense.scales": "model.safetensors",
393
+ "model.layers.21.self_attn.dense.weight": "model.safetensors",
394
+ "model.layers.21.self_attn.k_proj.bias": "model.safetensors",
395
+ "model.layers.21.self_attn.k_proj.biases": "model.safetensors",
396
+ "model.layers.21.self_attn.k_proj.scales": "model.safetensors",
397
+ "model.layers.21.self_attn.k_proj.weight": "model.safetensors",
398
+ "model.layers.21.self_attn.q_proj.bias": "model.safetensors",
399
+ "model.layers.21.self_attn.q_proj.biases": "model.safetensors",
400
+ "model.layers.21.self_attn.q_proj.scales": "model.safetensors",
401
+ "model.layers.21.self_attn.q_proj.weight": "model.safetensors",
402
+ "model.layers.21.self_attn.v_proj.bias": "model.safetensors",
403
+ "model.layers.21.self_attn.v_proj.biases": "model.safetensors",
404
+ "model.layers.21.self_attn.v_proj.scales": "model.safetensors",
405
+ "model.layers.21.self_attn.v_proj.weight": "model.safetensors",
406
+ "model.layers.22.input_layernorm.bias": "model.safetensors",
407
+ "model.layers.22.input_layernorm.weight": "model.safetensors",
408
+ "model.layers.22.mlp.fc1.bias": "model.safetensors",
409
+ "model.layers.22.mlp.fc1.biases": "model.safetensors",
410
+ "model.layers.22.mlp.fc1.scales": "model.safetensors",
411
+ "model.layers.22.mlp.fc1.weight": "model.safetensors",
412
+ "model.layers.22.mlp.fc2.bias": "model.safetensors",
413
+ "model.layers.22.mlp.fc2.biases": "model.safetensors",
414
+ "model.layers.22.mlp.fc2.scales": "model.safetensors",
415
+ "model.layers.22.mlp.fc2.weight": "model.safetensors",
416
+ "model.layers.22.self_attn.dense.bias": "model.safetensors",
417
+ "model.layers.22.self_attn.dense.biases": "model.safetensors",
418
+ "model.layers.22.self_attn.dense.scales": "model.safetensors",
419
+ "model.layers.22.self_attn.dense.weight": "model.safetensors",
420
+ "model.layers.22.self_attn.k_proj.bias": "model.safetensors",
421
+ "model.layers.22.self_attn.k_proj.biases": "model.safetensors",
422
+ "model.layers.22.self_attn.k_proj.scales": "model.safetensors",
423
+ "model.layers.22.self_attn.k_proj.weight": "model.safetensors",
424
+ "model.layers.22.self_attn.q_proj.bias": "model.safetensors",
425
+ "model.layers.22.self_attn.q_proj.biases": "model.safetensors",
426
+ "model.layers.22.self_attn.q_proj.scales": "model.safetensors",
427
+ "model.layers.22.self_attn.q_proj.weight": "model.safetensors",
428
+ "model.layers.22.self_attn.v_proj.bias": "model.safetensors",
429
+ "model.layers.22.self_attn.v_proj.biases": "model.safetensors",
430
+ "model.layers.22.self_attn.v_proj.scales": "model.safetensors",
431
+ "model.layers.22.self_attn.v_proj.weight": "model.safetensors",
432
+ "model.layers.23.input_layernorm.bias": "model.safetensors",
433
+ "model.layers.23.input_layernorm.weight": "model.safetensors",
434
+ "model.layers.23.mlp.fc1.bias": "model.safetensors",
435
+ "model.layers.23.mlp.fc1.biases": "model.safetensors",
436
+ "model.layers.23.mlp.fc1.scales": "model.safetensors",
437
+ "model.layers.23.mlp.fc1.weight": "model.safetensors",
438
+ "model.layers.23.mlp.fc2.bias": "model.safetensors",
439
+ "model.layers.23.mlp.fc2.biases": "model.safetensors",
440
+ "model.layers.23.mlp.fc2.scales": "model.safetensors",
441
+ "model.layers.23.mlp.fc2.weight": "model.safetensors",
442
+ "model.layers.23.self_attn.dense.bias": "model.safetensors",
443
+ "model.layers.23.self_attn.dense.biases": "model.safetensors",
444
+ "model.layers.23.self_attn.dense.scales": "model.safetensors",
445
+ "model.layers.23.self_attn.dense.weight": "model.safetensors",
446
+ "model.layers.23.self_attn.k_proj.bias": "model.safetensors",
447
+ "model.layers.23.self_attn.k_proj.biases": "model.safetensors",
448
+ "model.layers.23.self_attn.k_proj.scales": "model.safetensors",
449
+ "model.layers.23.self_attn.k_proj.weight": "model.safetensors",
450
+ "model.layers.23.self_attn.q_proj.bias": "model.safetensors",
451
+ "model.layers.23.self_attn.q_proj.biases": "model.safetensors",
452
+ "model.layers.23.self_attn.q_proj.scales": "model.safetensors",
453
+ "model.layers.23.self_attn.q_proj.weight": "model.safetensors",
454
+ "model.layers.23.self_attn.v_proj.bias": "model.safetensors",
455
+ "model.layers.23.self_attn.v_proj.biases": "model.safetensors",
456
+ "model.layers.23.self_attn.v_proj.scales": "model.safetensors",
457
+ "model.layers.23.self_attn.v_proj.weight": "model.safetensors",
458
+ "model.layers.24.input_layernorm.bias": "model.safetensors",
459
+ "model.layers.24.input_layernorm.weight": "model.safetensors",
460
+ "model.layers.24.mlp.fc1.bias": "model.safetensors",
461
+ "model.layers.24.mlp.fc1.biases": "model.safetensors",
462
+ "model.layers.24.mlp.fc1.scales": "model.safetensors",
463
+ "model.layers.24.mlp.fc1.weight": "model.safetensors",
464
+ "model.layers.24.mlp.fc2.bias": "model.safetensors",
465
+ "model.layers.24.mlp.fc2.biases": "model.safetensors",
466
+ "model.layers.24.mlp.fc2.scales": "model.safetensors",
467
+ "model.layers.24.mlp.fc2.weight": "model.safetensors",
468
+ "model.layers.24.self_attn.dense.bias": "model.safetensors",
469
+ "model.layers.24.self_attn.dense.biases": "model.safetensors",
470
+ "model.layers.24.self_attn.dense.scales": "model.safetensors",
471
+ "model.layers.24.self_attn.dense.weight": "model.safetensors",
472
+ "model.layers.24.self_attn.k_proj.bias": "model.safetensors",
473
+ "model.layers.24.self_attn.k_proj.biases": "model.safetensors",
474
+ "model.layers.24.self_attn.k_proj.scales": "model.safetensors",
475
+ "model.layers.24.self_attn.k_proj.weight": "model.safetensors",
476
+ "model.layers.24.self_attn.q_proj.bias": "model.safetensors",
477
+ "model.layers.24.self_attn.q_proj.biases": "model.safetensors",
478
+ "model.layers.24.self_attn.q_proj.scales": "model.safetensors",
479
+ "model.layers.24.self_attn.q_proj.weight": "model.safetensors",
480
+ "model.layers.24.self_attn.v_proj.bias": "model.safetensors",
481
+ "model.layers.24.self_attn.v_proj.biases": "model.safetensors",
482
+ "model.layers.24.self_attn.v_proj.scales": "model.safetensors",
483
+ "model.layers.24.self_attn.v_proj.weight": "model.safetensors",
484
+ "model.layers.25.input_layernorm.bias": "model.safetensors",
485
+ "model.layers.25.input_layernorm.weight": "model.safetensors",
486
+ "model.layers.25.mlp.fc1.bias": "model.safetensors",
487
+ "model.layers.25.mlp.fc1.biases": "model.safetensors",
488
+ "model.layers.25.mlp.fc1.scales": "model.safetensors",
489
+ "model.layers.25.mlp.fc1.weight": "model.safetensors",
490
+ "model.layers.25.mlp.fc2.bias": "model.safetensors",
491
+ "model.layers.25.mlp.fc2.biases": "model.safetensors",
492
+ "model.layers.25.mlp.fc2.scales": "model.safetensors",
493
+ "model.layers.25.mlp.fc2.weight": "model.safetensors",
494
+ "model.layers.25.self_attn.dense.bias": "model.safetensors",
495
+ "model.layers.25.self_attn.dense.biases": "model.safetensors",
496
+ "model.layers.25.self_attn.dense.scales": "model.safetensors",
497
+ "model.layers.25.self_attn.dense.weight": "model.safetensors",
498
+ "model.layers.25.self_attn.k_proj.bias": "model.safetensors",
499
+ "model.layers.25.self_attn.k_proj.biases": "model.safetensors",
500
+ "model.layers.25.self_attn.k_proj.scales": "model.safetensors",
501
+ "model.layers.25.self_attn.k_proj.weight": "model.safetensors",
502
+ "model.layers.25.self_attn.q_proj.bias": "model.safetensors",
503
+ "model.layers.25.self_attn.q_proj.biases": "model.safetensors",
504
+ "model.layers.25.self_attn.q_proj.scales": "model.safetensors",
505
+ "model.layers.25.self_attn.q_proj.weight": "model.safetensors",
506
+ "model.layers.25.self_attn.v_proj.bias": "model.safetensors",
507
+ "model.layers.25.self_attn.v_proj.biases": "model.safetensors",
508
+ "model.layers.25.self_attn.v_proj.scales": "model.safetensors",
509
+ "model.layers.25.self_attn.v_proj.weight": "model.safetensors",
510
+ "model.layers.26.input_layernorm.bias": "model.safetensors",
511
+ "model.layers.26.input_layernorm.weight": "model.safetensors",
512
+ "model.layers.26.mlp.fc1.bias": "model.safetensors",
513
+ "model.layers.26.mlp.fc1.biases": "model.safetensors",
514
+ "model.layers.26.mlp.fc1.scales": "model.safetensors",
515
+ "model.layers.26.mlp.fc1.weight": "model.safetensors",
516
+ "model.layers.26.mlp.fc2.bias": "model.safetensors",
517
+ "model.layers.26.mlp.fc2.biases": "model.safetensors",
518
+ "model.layers.26.mlp.fc2.scales": "model.safetensors",
519
+ "model.layers.26.mlp.fc2.weight": "model.safetensors",
520
+ "model.layers.26.self_attn.dense.bias": "model.safetensors",
521
+ "model.layers.26.self_attn.dense.biases": "model.safetensors",
522
+ "model.layers.26.self_attn.dense.scales": "model.safetensors",
523
+ "model.layers.26.self_attn.dense.weight": "model.safetensors",
524
+ "model.layers.26.self_attn.k_proj.bias": "model.safetensors",
525
+ "model.layers.26.self_attn.k_proj.biases": "model.safetensors",
526
+ "model.layers.26.self_attn.k_proj.scales": "model.safetensors",
527
+ "model.layers.26.self_attn.k_proj.weight": "model.safetensors",
528
+ "model.layers.26.self_attn.q_proj.bias": "model.safetensors",
529
+ "model.layers.26.self_attn.q_proj.biases": "model.safetensors",
530
+ "model.layers.26.self_attn.q_proj.scales": "model.safetensors",
531
+ "model.layers.26.self_attn.q_proj.weight": "model.safetensors",
532
+ "model.layers.26.self_attn.v_proj.bias": "model.safetensors",
533
+ "model.layers.26.self_attn.v_proj.biases": "model.safetensors",
534
+ "model.layers.26.self_attn.v_proj.scales": "model.safetensors",
535
+ "model.layers.26.self_attn.v_proj.weight": "model.safetensors",
536
+ "model.layers.27.input_layernorm.bias": "model.safetensors",
537
+ "model.layers.27.input_layernorm.weight": "model.safetensors",
538
+ "model.layers.27.mlp.fc1.bias": "model.safetensors",
539
+ "model.layers.27.mlp.fc1.biases": "model.safetensors",
540
+ "model.layers.27.mlp.fc1.scales": "model.safetensors",
541
+ "model.layers.27.mlp.fc1.weight": "model.safetensors",
542
+ "model.layers.27.mlp.fc2.bias": "model.safetensors",
543
+ "model.layers.27.mlp.fc2.biases": "model.safetensors",
544
+ "model.layers.27.mlp.fc2.scales": "model.safetensors",
545
+ "model.layers.27.mlp.fc2.weight": "model.safetensors",
546
+ "model.layers.27.self_attn.dense.bias": "model.safetensors",
547
+ "model.layers.27.self_attn.dense.biases": "model.safetensors",
548
+ "model.layers.27.self_attn.dense.scales": "model.safetensors",
549
+ "model.layers.27.self_attn.dense.weight": "model.safetensors",
550
+ "model.layers.27.self_attn.k_proj.bias": "model.safetensors",
551
+ "model.layers.27.self_attn.k_proj.biases": "model.safetensors",
552
+ "model.layers.27.self_attn.k_proj.scales": "model.safetensors",
553
+ "model.layers.27.self_attn.k_proj.weight": "model.safetensors",
554
+ "model.layers.27.self_attn.q_proj.bias": "model.safetensors",
555
+ "model.layers.27.self_attn.q_proj.biases": "model.safetensors",
556
+ "model.layers.27.self_attn.q_proj.scales": "model.safetensors",
557
+ "model.layers.27.self_attn.q_proj.weight": "model.safetensors",
558
+ "model.layers.27.self_attn.v_proj.bias": "model.safetensors",
559
+ "model.layers.27.self_attn.v_proj.biases": "model.safetensors",
560
+ "model.layers.27.self_attn.v_proj.scales": "model.safetensors",
561
+ "model.layers.27.self_attn.v_proj.weight": "model.safetensors",
562
+ "model.layers.28.input_layernorm.bias": "model.safetensors",
563
+ "model.layers.28.input_layernorm.weight": "model.safetensors",
564
+ "model.layers.28.mlp.fc1.bias": "model.safetensors",
565
+ "model.layers.28.mlp.fc1.biases": "model.safetensors",
566
+ "model.layers.28.mlp.fc1.scales": "model.safetensors",
567
+ "model.layers.28.mlp.fc1.weight": "model.safetensors",
568
+ "model.layers.28.mlp.fc2.bias": "model.safetensors",
569
+ "model.layers.28.mlp.fc2.biases": "model.safetensors",
570
+ "model.layers.28.mlp.fc2.scales": "model.safetensors",
571
+ "model.layers.28.mlp.fc2.weight": "model.safetensors",
572
+ "model.layers.28.self_attn.dense.bias": "model.safetensors",
573
+ "model.layers.28.self_attn.dense.biases": "model.safetensors",
574
+ "model.layers.28.self_attn.dense.scales": "model.safetensors",
575
+ "model.layers.28.self_attn.dense.weight": "model.safetensors",
576
+ "model.layers.28.self_attn.k_proj.bias": "model.safetensors",
577
+ "model.layers.28.self_attn.k_proj.biases": "model.safetensors",
578
+ "model.layers.28.self_attn.k_proj.scales": "model.safetensors",
579
+ "model.layers.28.self_attn.k_proj.weight": "model.safetensors",
580
+ "model.layers.28.self_attn.q_proj.bias": "model.safetensors",
581
+ "model.layers.28.self_attn.q_proj.biases": "model.safetensors",
582
+ "model.layers.28.self_attn.q_proj.scales": "model.safetensors",
583
+ "model.layers.28.self_attn.q_proj.weight": "model.safetensors",
584
+ "model.layers.28.self_attn.v_proj.bias": "model.safetensors",
585
+ "model.layers.28.self_attn.v_proj.biases": "model.safetensors",
586
+ "model.layers.28.self_attn.v_proj.scales": "model.safetensors",
587
+ "model.layers.28.self_attn.v_proj.weight": "model.safetensors",
588
+ "model.layers.29.input_layernorm.bias": "model.safetensors",
589
+ "model.layers.29.input_layernorm.weight": "model.safetensors",
590
+ "model.layers.29.mlp.fc1.bias": "model.safetensors",
591
+ "model.layers.29.mlp.fc1.biases": "model.safetensors",
592
+ "model.layers.29.mlp.fc1.scales": "model.safetensors",
593
+ "model.layers.29.mlp.fc1.weight": "model.safetensors",
594
+ "model.layers.29.mlp.fc2.bias": "model.safetensors",
595
+ "model.layers.29.mlp.fc2.biases": "model.safetensors",
596
+ "model.layers.29.mlp.fc2.scales": "model.safetensors",
597
+ "model.layers.29.mlp.fc2.weight": "model.safetensors",
598
+ "model.layers.29.self_attn.dense.bias": "model.safetensors",
599
+ "model.layers.29.self_attn.dense.biases": "model.safetensors",
600
+ "model.layers.29.self_attn.dense.scales": "model.safetensors",
601
+ "model.layers.29.self_attn.dense.weight": "model.safetensors",
602
+ "model.layers.29.self_attn.k_proj.bias": "model.safetensors",
603
+ "model.layers.29.self_attn.k_proj.biases": "model.safetensors",
604
+ "model.layers.29.self_attn.k_proj.scales": "model.safetensors",
605
+ "model.layers.29.self_attn.k_proj.weight": "model.safetensors",
606
+ "model.layers.29.self_attn.q_proj.bias": "model.safetensors",
607
+ "model.layers.29.self_attn.q_proj.biases": "model.safetensors",
608
+ "model.layers.29.self_attn.q_proj.scales": "model.safetensors",
609
+ "model.layers.29.self_attn.q_proj.weight": "model.safetensors",
610
+ "model.layers.29.self_attn.v_proj.bias": "model.safetensors",
611
+ "model.layers.29.self_attn.v_proj.biases": "model.safetensors",
612
+ "model.layers.29.self_attn.v_proj.scales": "model.safetensors",
613
+ "model.layers.29.self_attn.v_proj.weight": "model.safetensors",
614
+ "model.layers.3.input_layernorm.bias": "model.safetensors",
615
+ "model.layers.3.input_layernorm.weight": "model.safetensors",
616
+ "model.layers.3.mlp.fc1.bias": "model.safetensors",
617
+ "model.layers.3.mlp.fc1.biases": "model.safetensors",
618
+ "model.layers.3.mlp.fc1.scales": "model.safetensors",
619
+ "model.layers.3.mlp.fc1.weight": "model.safetensors",
620
+ "model.layers.3.mlp.fc2.bias": "model.safetensors",
621
+ "model.layers.3.mlp.fc2.biases": "model.safetensors",
622
+ "model.layers.3.mlp.fc2.scales": "model.safetensors",
623
+ "model.layers.3.mlp.fc2.weight": "model.safetensors",
624
+ "model.layers.3.self_attn.dense.bias": "model.safetensors",
625
+ "model.layers.3.self_attn.dense.biases": "model.safetensors",
626
+ "model.layers.3.self_attn.dense.scales": "model.safetensors",
627
+ "model.layers.3.self_attn.dense.weight": "model.safetensors",
628
+ "model.layers.3.self_attn.k_proj.bias": "model.safetensors",
629
+ "model.layers.3.self_attn.k_proj.biases": "model.safetensors",
630
+ "model.layers.3.self_attn.k_proj.scales": "model.safetensors",
631
+ "model.layers.3.self_attn.k_proj.weight": "model.safetensors",
632
+ "model.layers.3.self_attn.q_proj.bias": "model.safetensors",
633
+ "model.layers.3.self_attn.q_proj.biases": "model.safetensors",
634
+ "model.layers.3.self_attn.q_proj.scales": "model.safetensors",
635
+ "model.layers.3.self_attn.q_proj.weight": "model.safetensors",
636
+ "model.layers.3.self_attn.v_proj.bias": "model.safetensors",
637
+ "model.layers.3.self_attn.v_proj.biases": "model.safetensors",
638
+ "model.layers.3.self_attn.v_proj.scales": "model.safetensors",
639
+ "model.layers.3.self_attn.v_proj.weight": "model.safetensors",
640
+ "model.layers.30.input_layernorm.bias": "model.safetensors",
641
+ "model.layers.30.input_layernorm.weight": "model.safetensors",
642
+ "model.layers.30.mlp.fc1.bias": "model.safetensors",
643
+ "model.layers.30.mlp.fc1.biases": "model.safetensors",
644
+ "model.layers.30.mlp.fc1.scales": "model.safetensors",
645
+ "model.layers.30.mlp.fc1.weight": "model.safetensors",
646
+ "model.layers.30.mlp.fc2.bias": "model.safetensors",
647
+ "model.layers.30.mlp.fc2.biases": "model.safetensors",
648
+ "model.layers.30.mlp.fc2.scales": "model.safetensors",
649
+ "model.layers.30.mlp.fc2.weight": "model.safetensors",
650
+ "model.layers.30.self_attn.dense.bias": "model.safetensors",
651
+ "model.layers.30.self_attn.dense.biases": "model.safetensors",
652
+ "model.layers.30.self_attn.dense.scales": "model.safetensors",
653
+ "model.layers.30.self_attn.dense.weight": "model.safetensors",
654
+ "model.layers.30.self_attn.k_proj.bias": "model.safetensors",
655
+ "model.layers.30.self_attn.k_proj.biases": "model.safetensors",
656
+ "model.layers.30.self_attn.k_proj.scales": "model.safetensors",
657
+ "model.layers.30.self_attn.k_proj.weight": "model.safetensors",
658
+ "model.layers.30.self_attn.q_proj.bias": "model.safetensors",
659
+ "model.layers.30.self_attn.q_proj.biases": "model.safetensors",
660
+ "model.layers.30.self_attn.q_proj.scales": "model.safetensors",
661
+ "model.layers.30.self_attn.q_proj.weight": "model.safetensors",
662
+ "model.layers.30.self_attn.v_proj.bias": "model.safetensors",
663
+ "model.layers.30.self_attn.v_proj.biases": "model.safetensors",
664
+ "model.layers.30.self_attn.v_proj.scales": "model.safetensors",
665
+ "model.layers.30.self_attn.v_proj.weight": "model.safetensors",
666
+ "model.layers.31.input_layernorm.bias": "model.safetensors",
667
+ "model.layers.31.input_layernorm.weight": "model.safetensors",
668
+ "model.layers.31.mlp.fc1.bias": "model.safetensors",
669
+ "model.layers.31.mlp.fc1.biases": "model.safetensors",
670
+ "model.layers.31.mlp.fc1.scales": "model.safetensors",
671
+ "model.layers.31.mlp.fc1.weight": "model.safetensors",
672
+ "model.layers.31.mlp.fc2.bias": "model.safetensors",
673
+ "model.layers.31.mlp.fc2.biases": "model.safetensors",
674
+ "model.layers.31.mlp.fc2.scales": "model.safetensors",
675
+ "model.layers.31.mlp.fc2.weight": "model.safetensors",
676
+ "model.layers.31.self_attn.dense.bias": "model.safetensors",
677
+ "model.layers.31.self_attn.dense.biases": "model.safetensors",
678
+ "model.layers.31.self_attn.dense.scales": "model.safetensors",
679
+ "model.layers.31.self_attn.dense.weight": "model.safetensors",
680
+ "model.layers.31.self_attn.k_proj.bias": "model.safetensors",
681
+ "model.layers.31.self_attn.k_proj.biases": "model.safetensors",
682
+ "model.layers.31.self_attn.k_proj.scales": "model.safetensors",
683
+ "model.layers.31.self_attn.k_proj.weight": "model.safetensors",
684
+ "model.layers.31.self_attn.q_proj.bias": "model.safetensors",
685
+ "model.layers.31.self_attn.q_proj.biases": "model.safetensors",
686
+ "model.layers.31.self_attn.q_proj.scales": "model.safetensors",
687
+ "model.layers.31.self_attn.q_proj.weight": "model.safetensors",
688
+ "model.layers.31.self_attn.v_proj.bias": "model.safetensors",
689
+ "model.layers.31.self_attn.v_proj.biases": "model.safetensors",
690
+ "model.layers.31.self_attn.v_proj.scales": "model.safetensors",
691
+ "model.layers.31.self_attn.v_proj.weight": "model.safetensors",
692
+ "model.layers.4.input_layernorm.bias": "model.safetensors",
693
+ "model.layers.4.input_layernorm.weight": "model.safetensors",
694
+ "model.layers.4.mlp.fc1.bias": "model.safetensors",
695
+ "model.layers.4.mlp.fc1.biases": "model.safetensors",
696
+ "model.layers.4.mlp.fc1.scales": "model.safetensors",
697
+ "model.layers.4.mlp.fc1.weight": "model.safetensors",
698
+ "model.layers.4.mlp.fc2.bias": "model.safetensors",
699
+ "model.layers.4.mlp.fc2.biases": "model.safetensors",
700
+ "model.layers.4.mlp.fc2.scales": "model.safetensors",
701
+ "model.layers.4.mlp.fc2.weight": "model.safetensors",
702
+ "model.layers.4.self_attn.dense.bias": "model.safetensors",
703
+ "model.layers.4.self_attn.dense.biases": "model.safetensors",
704
+ "model.layers.4.self_attn.dense.scales": "model.safetensors",
705
+ "model.layers.4.self_attn.dense.weight": "model.safetensors",
706
+ "model.layers.4.self_attn.k_proj.bias": "model.safetensors",
707
+ "model.layers.4.self_attn.k_proj.biases": "model.safetensors",
708
+ "model.layers.4.self_attn.k_proj.scales": "model.safetensors",
709
+ "model.layers.4.self_attn.k_proj.weight": "model.safetensors",
710
+ "model.layers.4.self_attn.q_proj.bias": "model.safetensors",
711
+ "model.layers.4.self_attn.q_proj.biases": "model.safetensors",
712
+ "model.layers.4.self_attn.q_proj.scales": "model.safetensors",
713
+ "model.layers.4.self_attn.q_proj.weight": "model.safetensors",
714
+ "model.layers.4.self_attn.v_proj.bias": "model.safetensors",
715
+ "model.layers.4.self_attn.v_proj.biases": "model.safetensors",
716
+ "model.layers.4.self_attn.v_proj.scales": "model.safetensors",
717
+ "model.layers.4.self_attn.v_proj.weight": "model.safetensors",
718
+ "model.layers.5.input_layernorm.bias": "model.safetensors",
719
+ "model.layers.5.input_layernorm.weight": "model.safetensors",
720
+ "model.layers.5.mlp.fc1.bias": "model.safetensors",
721
+ "model.layers.5.mlp.fc1.biases": "model.safetensors",
722
+ "model.layers.5.mlp.fc1.scales": "model.safetensors",
723
+ "model.layers.5.mlp.fc1.weight": "model.safetensors",
724
+ "model.layers.5.mlp.fc2.bias": "model.safetensors",
725
+ "model.layers.5.mlp.fc2.biases": "model.safetensors",
726
+ "model.layers.5.mlp.fc2.scales": "model.safetensors",
727
+ "model.layers.5.mlp.fc2.weight": "model.safetensors",
728
+ "model.layers.5.self_attn.dense.bias": "model.safetensors",
729
+ "model.layers.5.self_attn.dense.biases": "model.safetensors",
730
+ "model.layers.5.self_attn.dense.scales": "model.safetensors",
731
+ "model.layers.5.self_attn.dense.weight": "model.safetensors",
732
+ "model.layers.5.self_attn.k_proj.bias": "model.safetensors",
733
+ "model.layers.5.self_attn.k_proj.biases": "model.safetensors",
734
+ "model.layers.5.self_attn.k_proj.scales": "model.safetensors",
735
+ "model.layers.5.self_attn.k_proj.weight": "model.safetensors",
736
+ "model.layers.5.self_attn.q_proj.bias": "model.safetensors",
737
+ "model.layers.5.self_attn.q_proj.biases": "model.safetensors",
738
+ "model.layers.5.self_attn.q_proj.scales": "model.safetensors",
739
+ "model.layers.5.self_attn.q_proj.weight": "model.safetensors",
740
+ "model.layers.5.self_attn.v_proj.bias": "model.safetensors",
741
+ "model.layers.5.self_attn.v_proj.biases": "model.safetensors",
742
+ "model.layers.5.self_attn.v_proj.scales": "model.safetensors",
743
+ "model.layers.5.self_attn.v_proj.weight": "model.safetensors",
744
+ "model.layers.6.input_layernorm.bias": "model.safetensors",
745
+ "model.layers.6.input_layernorm.weight": "model.safetensors",
746
+ "model.layers.6.mlp.fc1.bias": "model.safetensors",
747
+ "model.layers.6.mlp.fc1.biases": "model.safetensors",
748
+ "model.layers.6.mlp.fc1.scales": "model.safetensors",
749
+ "model.layers.6.mlp.fc1.weight": "model.safetensors",
750
+ "model.layers.6.mlp.fc2.bias": "model.safetensors",
751
+ "model.layers.6.mlp.fc2.biases": "model.safetensors",
752
+ "model.layers.6.mlp.fc2.scales": "model.safetensors",
753
+ "model.layers.6.mlp.fc2.weight": "model.safetensors",
754
+ "model.layers.6.self_attn.dense.bias": "model.safetensors",
755
+ "model.layers.6.self_attn.dense.biases": "model.safetensors",
756
+ "model.layers.6.self_attn.dense.scales": "model.safetensors",
757
+ "model.layers.6.self_attn.dense.weight": "model.safetensors",
758
+ "model.layers.6.self_attn.k_proj.bias": "model.safetensors",
759
+ "model.layers.6.self_attn.k_proj.biases": "model.safetensors",
760
+ "model.layers.6.self_attn.k_proj.scales": "model.safetensors",
761
+ "model.layers.6.self_attn.k_proj.weight": "model.safetensors",
762
+ "model.layers.6.self_attn.q_proj.bias": "model.safetensors",
763
+ "model.layers.6.self_attn.q_proj.biases": "model.safetensors",
764
+ "model.layers.6.self_attn.q_proj.scales": "model.safetensors",
765
+ "model.layers.6.self_attn.q_proj.weight": "model.safetensors",
766
+ "model.layers.6.self_attn.v_proj.bias": "model.safetensors",
767
+ "model.layers.6.self_attn.v_proj.biases": "model.safetensors",
768
+ "model.layers.6.self_attn.v_proj.scales": "model.safetensors",
769
+ "model.layers.6.self_attn.v_proj.weight": "model.safetensors",
770
+ "model.layers.7.input_layernorm.bias": "model.safetensors",
771
+ "model.layers.7.input_layernorm.weight": "model.safetensors",
772
+ "model.layers.7.mlp.fc1.bias": "model.safetensors",
773
+ "model.layers.7.mlp.fc1.biases": "model.safetensors",
774
+ "model.layers.7.mlp.fc1.scales": "model.safetensors",
775
+ "model.layers.7.mlp.fc1.weight": "model.safetensors",
776
+ "model.layers.7.mlp.fc2.bias": "model.safetensors",
777
+ "model.layers.7.mlp.fc2.biases": "model.safetensors",
778
+ "model.layers.7.mlp.fc2.scales": "model.safetensors",
779
+ "model.layers.7.mlp.fc2.weight": "model.safetensors",
780
+ "model.layers.7.self_attn.dense.bias": "model.safetensors",
781
+ "model.layers.7.self_attn.dense.biases": "model.safetensors",
782
+ "model.layers.7.self_attn.dense.scales": "model.safetensors",
783
+ "model.layers.7.self_attn.dense.weight": "model.safetensors",
784
+ "model.layers.7.self_attn.k_proj.bias": "model.safetensors",
785
+ "model.layers.7.self_attn.k_proj.biases": "model.safetensors",
786
+ "model.layers.7.self_attn.k_proj.scales": "model.safetensors",
787
+ "model.layers.7.self_attn.k_proj.weight": "model.safetensors",
788
+ "model.layers.7.self_attn.q_proj.bias": "model.safetensors",
789
+ "model.layers.7.self_attn.q_proj.biases": "model.safetensors",
790
+ "model.layers.7.self_attn.q_proj.scales": "model.safetensors",
791
+ "model.layers.7.self_attn.q_proj.weight": "model.safetensors",
792
+ "model.layers.7.self_attn.v_proj.bias": "model.safetensors",
793
+ "model.layers.7.self_attn.v_proj.biases": "model.safetensors",
794
+ "model.layers.7.self_attn.v_proj.scales": "model.safetensors",
795
+ "model.layers.7.self_attn.v_proj.weight": "model.safetensors",
796
+ "model.layers.8.input_layernorm.bias": "model.safetensors",
797
+ "model.layers.8.input_layernorm.weight": "model.safetensors",
798
+ "model.layers.8.mlp.fc1.bias": "model.safetensors",
799
+ "model.layers.8.mlp.fc1.biases": "model.safetensors",
800
+ "model.layers.8.mlp.fc1.scales": "model.safetensors",
801
+ "model.layers.8.mlp.fc1.weight": "model.safetensors",
802
+ "model.layers.8.mlp.fc2.bias": "model.safetensors",
803
+ "model.layers.8.mlp.fc2.biases": "model.safetensors",
804
+ "model.layers.8.mlp.fc2.scales": "model.safetensors",
805
+ "model.layers.8.mlp.fc2.weight": "model.safetensors",
806
+ "model.layers.8.self_attn.dense.bias": "model.safetensors",
807
+ "model.layers.8.self_attn.dense.biases": "model.safetensors",
808
+ "model.layers.8.self_attn.dense.scales": "model.safetensors",
809
+ "model.layers.8.self_attn.dense.weight": "model.safetensors",
810
+ "model.layers.8.self_attn.k_proj.bias": "model.safetensors",
811
+ "model.layers.8.self_attn.k_proj.biases": "model.safetensors",
812
+ "model.layers.8.self_attn.k_proj.scales": "model.safetensors",
813
+ "model.layers.8.self_attn.k_proj.weight": "model.safetensors",
814
+ "model.layers.8.self_attn.q_proj.bias": "model.safetensors",
815
+ "model.layers.8.self_attn.q_proj.biases": "model.safetensors",
816
+ "model.layers.8.self_attn.q_proj.scales": "model.safetensors",
817
+ "model.layers.8.self_attn.q_proj.weight": "model.safetensors",
818
+ "model.layers.8.self_attn.v_proj.bias": "model.safetensors",
819
+ "model.layers.8.self_attn.v_proj.biases": "model.safetensors",
820
+ "model.layers.8.self_attn.v_proj.scales": "model.safetensors",
821
+ "model.layers.8.self_attn.v_proj.weight": "model.safetensors",
822
+ "model.layers.9.input_layernorm.bias": "model.safetensors",
823
+ "model.layers.9.input_layernorm.weight": "model.safetensors",
824
+ "model.layers.9.mlp.fc1.bias": "model.safetensors",
825
+ "model.layers.9.mlp.fc1.biases": "model.safetensors",
826
+ "model.layers.9.mlp.fc1.scales": "model.safetensors",
827
+ "model.layers.9.mlp.fc1.weight": "model.safetensors",
828
+ "model.layers.9.mlp.fc2.bias": "model.safetensors",
829
+ "model.layers.9.mlp.fc2.biases": "model.safetensors",
830
+ "model.layers.9.mlp.fc2.scales": "model.safetensors",
831
+ "model.layers.9.mlp.fc2.weight": "model.safetensors",
832
+ "model.layers.9.self_attn.dense.bias": "model.safetensors",
833
+ "model.layers.9.self_attn.dense.biases": "model.safetensors",
834
+ "model.layers.9.self_attn.dense.scales": "model.safetensors",
835
+ "model.layers.9.self_attn.dense.weight": "model.safetensors",
836
+ "model.layers.9.self_attn.k_proj.bias": "model.safetensors",
837
+ "model.layers.9.self_attn.k_proj.biases": "model.safetensors",
838
+ "model.layers.9.self_attn.k_proj.scales": "model.safetensors",
839
+ "model.layers.9.self_attn.k_proj.weight": "model.safetensors",
840
+ "model.layers.9.self_attn.q_proj.bias": "model.safetensors",
841
+ "model.layers.9.self_attn.q_proj.biases": "model.safetensors",
842
+ "model.layers.9.self_attn.q_proj.scales": "model.safetensors",
843
+ "model.layers.9.self_attn.q_proj.weight": "model.safetensors",
844
+ "model.layers.9.self_attn.v_proj.bias": "model.safetensors",
845
+ "model.layers.9.self_attn.v_proj.biases": "model.safetensors",
846
+ "model.layers.9.self_attn.v_proj.scales": "model.safetensors",
847
+ "model.layers.9.self_attn.v_proj.weight": "model.safetensors"
848
+ }
849
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|endoftext|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "unk_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,328 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "50257": {
13
+ "content": " ",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": false
19
+ },
20
+ "50258": {
21
+ "content": " ",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": false
27
+ },
28
+ "50259": {
29
+ "content": " ",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": false
35
+ },
36
+ "50260": {
37
+ "content": " ",
38
+ "lstrip": false,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": false
43
+ },
44
+ "50261": {
45
+ "content": " ",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": false
51
+ },
52
+ "50262": {
53
+ "content": " ",
54
+ "lstrip": false,
55
+ "normalized": true,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": false
59
+ },
60
+ "50263": {
61
+ "content": " ",
62
+ "lstrip": false,
63
+ "normalized": true,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": false
67
+ },
68
+ "50264": {
69
+ "content": " ",
70
+ "lstrip": false,
71
+ "normalized": true,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": false
75
+ },
76
+ "50265": {
77
+ "content": " ",
78
+ "lstrip": false,
79
+ "normalized": true,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": false
83
+ },
84
+ "50266": {
85
+ "content": " ",
86
+ "lstrip": false,
87
+ "normalized": true,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": false
91
+ },
92
+ "50267": {
93
+ "content": " ",
94
+ "lstrip": false,
95
+ "normalized": true,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": false
99
+ },
100
+ "50268": {
101
+ "content": " ",
102
+ "lstrip": false,
103
+ "normalized": true,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": false
107
+ },
108
+ "50269": {
109
+ "content": " ",
110
+ "lstrip": false,
111
+ "normalized": true,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": false
115
+ },
116
+ "50270": {
117
+ "content": " ",
118
+ "lstrip": false,
119
+ "normalized": true,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "50271": {
125
+ "content": " ",
126
+ "lstrip": false,
127
+ "normalized": true,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "50272": {
133
+ "content": " ",
134
+ "lstrip": false,
135
+ "normalized": true,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "50273": {
141
+ "content": " ",
142
+ "lstrip": false,
143
+ "normalized": true,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "50274": {
149
+ "content": " ",
150
+ "lstrip": false,
151
+ "normalized": true,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "50275": {
157
+ "content": " ",
158
+ "lstrip": false,
159
+ "normalized": true,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "50276": {
165
+ "content": " ",
166
+ "lstrip": false,
167
+ "normalized": true,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "50277": {
173
+ "content": " ",
174
+ "lstrip": false,
175
+ "normalized": true,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "50278": {
181
+ "content": " ",
182
+ "lstrip": false,
183
+ "normalized": true,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "50279": {
189
+ "content": " ",
190
+ "lstrip": false,
191
+ "normalized": true,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "50280": {
197
+ "content": " ",
198
+ "lstrip": false,
199
+ "normalized": true,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "50281": {
205
+ "content": " ",
206
+ "lstrip": false,
207
+ "normalized": true,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ },
212
+ "50282": {
213
+ "content": " ",
214
+ "lstrip": false,
215
+ "normalized": true,
216
+ "rstrip": false,
217
+ "single_word": false,
218
+ "special": false
219
+ },
220
+ "50283": {
221
+ "content": " ",
222
+ "lstrip": false,
223
+ "normalized": true,
224
+ "rstrip": false,
225
+ "single_word": false,
226
+ "special": false
227
+ },
228
+ "50284": {
229
+ "content": " ",
230
+ "lstrip": false,
231
+ "normalized": true,
232
+ "rstrip": false,
233
+ "single_word": false,
234
+ "special": false
235
+ },
236
+ "50285": {
237
+ "content": " ",
238
+ "lstrip": false,
239
+ "normalized": true,
240
+ "rstrip": false,
241
+ "single_word": false,
242
+ "special": false
243
+ },
244
+ "50286": {
245
+ "content": " ",
246
+ "lstrip": false,
247
+ "normalized": true,
248
+ "rstrip": false,
249
+ "single_word": false,
250
+ "special": false
251
+ },
252
+ "50287": {
253
+ "content": "\t\t\t\t\t\t\t\t\t",
254
+ "lstrip": false,
255
+ "normalized": true,
256
+ "rstrip": false,
257
+ "single_word": false,
258
+ "special": false
259
+ },
260
+ "50288": {
261
+ "content": "\t\t\t\t\t\t\t\t",
262
+ "lstrip": false,
263
+ "normalized": true,
264
+ "rstrip": false,
265
+ "single_word": false,
266
+ "special": false
267
+ },
268
+ "50289": {
269
+ "content": "\t\t\t\t\t\t\t",
270
+ "lstrip": false,
271
+ "normalized": true,
272
+ "rstrip": false,
273
+ "single_word": false,
274
+ "special": false
275
+ },
276
+ "50290": {
277
+ "content": "\t\t\t\t\t\t",
278
+ "lstrip": false,
279
+ "normalized": true,
280
+ "rstrip": false,
281
+ "single_word": false,
282
+ "special": false
283
+ },
284
+ "50291": {
285
+ "content": "\t\t\t\t\t",
286
+ "lstrip": false,
287
+ "normalized": true,
288
+ "rstrip": false,
289
+ "single_word": false,
290
+ "special": false
291
+ },
292
+ "50292": {
293
+ "content": "\t\t\t\t",
294
+ "lstrip": false,
295
+ "normalized": true,
296
+ "rstrip": false,
297
+ "single_word": false,
298
+ "special": false
299
+ },
300
+ "50293": {
301
+ "content": "\t\t\t",
302
+ "lstrip": false,
303
+ "normalized": true,
304
+ "rstrip": false,
305
+ "single_word": false,
306
+ "special": false
307
+ },
308
+ "50294": {
309
+ "content": "\t\t",
310
+ "lstrip": false,
311
+ "normalized": true,
312
+ "rstrip": false,
313
+ "single_word": false,
314
+ "special": false
315
+ }
316
+ },
317
+ "backend": "tokenizers",
318
+ "bos_token": "<|endoftext|>",
319
+ "clean_up_tokenization_spaces": true,
320
+ "eos_token": "<|endoftext|>",
321
+ "extra_special_tokens": {},
322
+ "is_local": false,
323
+ "model_max_length": 4096,
324
+ "return_token_type_ids": false,
325
+ "tokenizer_class": "CodeGenTokenizer",
326
+ "unk_token": "<|endoftext|>",
327
+ "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{ message['content'] + '\n' }}{% elif message['role'] == 'user' %}### Instruction:\n{{ message['content'] }}\n\n### Response:\n{% elif message['role'] == 'assistant' %}{{ message['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}### Response:\n{% endif %}"
328
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
webicoder_icon.png ADDED

Git LFS Details

  • SHA256: 343a0976e565c05162129f1fb612f2a996156ec722b7fd824cba4560c6b2a02b
  • Pointer size: 131 Bytes
  • Size of remote file: 318 kB