Hugging Face's logo

zeromodels
/
gemma-7b-it

Text Generation
KerasFormers
Keras
PyTorch
JAX
TensorFlow
English
gemma
gemma-7b
Model card Files
xet
 Community
IMvision12 commited on
Commit
d9604ae
·
0 Parent(s):

Super-squash branch 'main' using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +36 -0
  2. README.md +80 -0
  3. kf_config.json +23 -0
  4. model.weights.json +769 -0
  5. model_00000.weights.h5 +3 -0
  6. model_00001.weights.h5 +3 -0
  7. model_00002.weights.h5 +3 -0
  8. model_00003.weights.h5 +3 -0
  9. tokenizer.json +3 -0
.gitattributes ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz 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 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pipeline_tag: text-generation
3
+ license: gemma
4
+ base_model: google/gemma-7b-it
5
+ library_name: kerasformers
6
+ extra_gated_heading: Access Gemma on Hugging Face
7
+ language:
8
+ - en
9
+ tags:
10
+ - keras
11
+ - kerasformers
12
+ - gemma
13
+ - gemma-7b
14
+ - text-generation
15
+ - arxiv:2403.08295
16
+ - pytorch
17
+ - jax
18
+ - tf
19
+ ---
20
+
21
+ ## ***See [our collection](https://huggingface.co/collections/kerasformers/gemma-6a69aaecae0f1f518733ffa1) for all versions of Gemma.***
22
+
23
+ # Run Gemma with Keras 3: JAX, PyTorch, or TensorFlow
24
+
25
+ [![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-Gemma-blue)](https://imvision12.github.io/KerasFormers/gemma/) [![Collection](https://img.shields.io/badge/HF-Gemma%20collection-yellow)](https://huggingface.co/collections/kerasformers/gemma-6a69aaecae0f1f518733ffa1)
26
+
27
+ # kerasformers/gemma-7b-it
28
+
29
+ Paper: [Gemma: Open Models Based on Gemini Research and Technology (arXiv:2403.08295)](https://arxiv.org/abs/2403.08295) · [HF Papers](https://huggingface.co/papers/2403.08295)
30
+
31
+ Gemma is Google's open decoder-only LLM family (RMSNorm, GeGLU, RoPE, multi-query attention). Base checkpoints are for completion; `-it` / `1.1` variants are instruction-tuned for chat.
32
+
33
+ For more details on the model, please go to Google's original [model card](https://huggingface.co/google/gemma-7b-it).
34
+
35
+ Pure-**Keras 3** conversion of [`google/gemma-7b-it`](https://huggingface.co/google/gemma-7b-it) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
36
+
37
+ This is an **instruction-tuned** checkpoint: use the chat template via `GemmaTokenizer`.
38
+
39
+ ## ✨ Quick start
40
+
41
+ ```python
42
+ import os
43
+ os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
44
+
45
+ from kerasformers.models.gemma import GemmaTextGenerate, GemmaTokenizer
46
+
47
+ model = GemmaTextGenerate.from_weights("kerasformers/gemma-7b-it")
48
+ tokenizer = GemmaTokenizer.from_weights("kerasformers/gemma-7b-it")
49
+
50
+ inputs = tokenizer([
51
+ {"role": "user", "content": "Explain rotary embeddings in one sentence."}
52
+ ])
53
+ outputs = model.generate(**inputs, max_new_tokens=64)
54
+ print(tokenizer.decode(outputs[0]))
55
+ ```
56
+
57
+ Load any Gemma v1 variant the same way with `from_weights("kerasformers/<variant>")`:
58
+
59
+ | Variant | Hub | Type |
60
+ |---|---|---|
61
+ | `gemma-2b` | [`kerasformers/gemma-2b`](https://huggingface.co/kerasformers/gemma-2b) | base |
62
+ | `gemma-2b-it` | [`kerasformers/gemma-2b-it`](https://huggingface.co/kerasformers/gemma-2b-it) | instruct |
63
+ | `gemma-1.1-2b-it` | [`kerasformers/gemma-1.1-2b-it`](https://huggingface.co/kerasformers/gemma-1.1-2b-it) | instruct (1.1) |
64
+ | `gemma-7b` | [`kerasformers/gemma-7b`](https://huggingface.co/kerasformers/gemma-7b) | base |
65
+ | `gemma-7b-it` | [`kerasformers/gemma-7b-it`](https://huggingface.co/kerasformers/gemma-7b-it) | instruct |
66
+ | `gemma-1.1-7b-it` | [`kerasformers/gemma-1.1-7b-it`](https://huggingface.co/kerasformers/gemma-1.1-7b-it) | instruct (1.1) |
67
+
68
+ ## Tips
69
+
70
+ - Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
71
+ - Prefer `GemmaTokenizer.from_weights(...)` so the chat template matches.
72
+ - Larger checkpoints: try `load_dtype="bfloat16"` or `quantization="int8"`.
73
+ - See [Gemma docs](https://imvision12.github.io/KerasFormers/gemma/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
74
+ - Community / upstream safetensors still work via the `hf:` prefix, e.g. `GemmaTextGenerate.from_weights("hf:google/gemma-7b-it")`.
75
+
76
+ ## Special Thanks
77
+
78
+ A huge thank you to the Google Gemma authors for creating and releasing these models.
79
+
80
+ License: Gemma (gated). Accept the license on the upstream Hub card before downloading.
kf_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "library_name": "kerasformers",
3
+ "kerasformers_version": "1.2.1",
4
+ "model_module": "kerasformers.models.gemma",
5
+ "model_class": "GemmaTextGenerate",
6
+ "variant": "gemma-7b-it",
7
+ "weights": "model.weights.json",
8
+ "schema_version": 2,
9
+ "weight_dtype": "bfloat16",
10
+ "model_type": "gemma",
11
+ "text_config": {
12
+ "vocab_size": 256000,
13
+ "embed_dim": 3072,
14
+ "mlp_dim": 24576,
15
+ "num_layers": 28,
16
+ "num_heads": 16,
17
+ "num_kv_heads": 16,
18
+ "head_dim": 256,
19
+ "norm_eps": 1e-06,
20
+ "rope_theta": 10000.0,
21
+ "tie_embeddings": true
22
+ }
23
+ }
model.weights.json ADDED
@@ -0,0 +1,769 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 17075361792.0
4
+ },
5
+ "weight_map": {
6
+ "/layers/embedding/vars": [
7
+ "model_00000.weights.h5"
8
+ ],
9
+ "/layers/gemma_decoder_layer/attention_norm/vars": [
10
+ "model_00000.weights.h5"
11
+ ],
12
+ "/layers/gemma_decoder_layer/mlp/down/vars": [
13
+ "model_00000.weights.h5"
14
+ ],
15
+ "/layers/gemma_decoder_layer/mlp/gate/vars": [
16
+ "model_00000.weights.h5"
17
+ ],
18
+ "/layers/gemma_decoder_layer/mlp/up/vars": [
19
+ "model_00000.weights.h5"
20
+ ],
21
+ "/layers/gemma_decoder_layer/mlp_norm/vars": [
22
+ "model_00000.weights.h5"
23
+ ],
24
+ "/layers/gemma_decoder_layer/attention/key/vars": [
25
+ "model_00000.weights.h5"
26
+ ],
27
+ "/layers/gemma_decoder_layer/attention/output_proj/vars": [
28
+ "model_00000.weights.h5"
29
+ ],
30
+ "/layers/gemma_decoder_layer/attention/query/vars": [
31
+ "model_00000.weights.h5"
32
+ ],
33
+ "/layers/gemma_decoder_layer/attention/value/vars": [
34
+ "model_00000.weights.h5"
35
+ ],
36
+ "/layers/gemma_decoder_layer_1/attention_norm/vars": [
37
+ "model_00000.weights.h5"
38
+ ],
39
+ "/layers/gemma_decoder_layer_1/mlp/down/vars": [
40
+ "model_00000.weights.h5"
41
+ ],
42
+ "/layers/gemma_decoder_layer_1/mlp/gate/vars": [
43
+ "model_00000.weights.h5"
44
+ ],
45
+ "/layers/gemma_decoder_layer_1/mlp/up/vars": [
46
+ "model_00000.weights.h5"
47
+ ],
48
+ "/layers/gemma_decoder_layer_1/mlp_norm/vars": [
49
+ "model_00000.weights.h5"
50
+ ],
51
+ "/layers/gemma_decoder_layer_1/attention/key/vars": [
52
+ "model_00000.weights.h5"
53
+ ],
54
+ "/layers/gemma_decoder_layer_1/attention/output_proj/vars": [
55
+ "model_00000.weights.h5"
56
+ ],
57
+ "/layers/gemma_decoder_layer_1/attention/query/vars": [
58
+ "model_00000.weights.h5"
59
+ ],
60
+ "/layers/gemma_decoder_layer_1/attention/value/vars": [
61
+ "model_00000.weights.h5"
62
+ ],
63
+ "/layers/gemma_decoder_layer_2/attention_norm/vars": [
64
+ "model_00000.weights.h5"
65
+ ],
66
+ "/layers/gemma_decoder_layer_2/mlp/down/vars": [
67
+ "model_00000.weights.h5"
68
+ ],
69
+ "/layers/gemma_decoder_layer_2/mlp/gate/vars": [
70
+ "model_00000.weights.h5"
71
+ ],
72
+ "/layers/gemma_decoder_layer_2/mlp/up/vars": [
73
+ "model_00000.weights.h5"
74
+ ],
75
+ "/layers/gemma_decoder_layer_2/mlp_norm/vars": [
76
+ "model_00000.weights.h5"
77
+ ],
78
+ "/layers/gemma_decoder_layer_2/attention/key/vars": [
79
+ "model_00000.weights.h5"
80
+ ],
81
+ "/layers/gemma_decoder_layer_2/attention/output_proj/vars": [
82
+ "model_00000.weights.h5"
83
+ ],
84
+ "/layers/gemma_decoder_layer_2/attention/query/vars": [
85
+ "model_00000.weights.h5"
86
+ ],
87
+ "/layers/gemma_decoder_layer_2/attention/value/vars": [
88
+ "model_00000.weights.h5"
89
+ ],
90
+ "/layers/gemma_decoder_layer_3/attention_norm/vars": [
91
+ "model_00000.weights.h5"
92
+ ],
93
+ "/layers/gemma_decoder_layer_3/mlp/down/vars": [
94
+ "model_00000.weights.h5"
95
+ ],
96
+ "/layers/gemma_decoder_layer_3/mlp/gate/vars": [
97
+ "model_00000.weights.h5"
98
+ ],
99
+ "/layers/gemma_decoder_layer_3/mlp/up/vars": [
100
+ "model_00000.weights.h5"
101
+ ],
102
+ "/layers/gemma_decoder_layer_3/mlp_norm/vars": [
103
+ "model_00000.weights.h5"
104
+ ],
105
+ "/layers/gemma_decoder_layer_3/attention/key/vars": [
106
+ "model_00000.weights.h5"
107
+ ],
108
+ "/layers/gemma_decoder_layer_3/attention/output_proj/vars": [
109
+ "model_00000.weights.h5"
110
+ ],
111
+ "/layers/gemma_decoder_layer_3/attention/query/vars": [
112
+ "model_00000.weights.h5"
113
+ ],
114
+ "/layers/gemma_decoder_layer_3/attention/value/vars": [
115
+ "model_00000.weights.h5"
116
+ ],
117
+ "/layers/gemma_decoder_layer_4/attention_norm/vars": [
118
+ "model_00000.weights.h5"
119
+ ],
120
+ "/layers/gemma_decoder_layer_4/mlp/down/vars": [
121
+ "model_00000.weights.h5"
122
+ ],
123
+ "/layers/gemma_decoder_layer_4/mlp/gate/vars": [
124
+ "model_00000.weights.h5"
125
+ ],
126
+ "/layers/gemma_decoder_layer_4/mlp/up/vars": [
127
+ "model_00000.weights.h5"
128
+ ],
129
+ "/layers/gemma_decoder_layer_4/mlp_norm/vars": [
130
+ "model_00000.weights.h5"
131
+ ],
132
+ "/layers/gemma_decoder_layer_4/attention/key/vars": [
133
+ "model_00000.weights.h5"
134
+ ],
135
+ "/layers/gemma_decoder_layer_4/attention/output_proj/vars": [
136
+ "model_00000.weights.h5"
137
+ ],
138
+ "/layers/gemma_decoder_layer_4/attention/query/vars": [
139
+ "model_00000.weights.h5"
140
+ ],
141
+ "/layers/gemma_decoder_layer_4/attention/value/vars": [
142
+ "model_00000.weights.h5"
143
+ ],
144
+ "/layers/gemma_decoder_layer_5/attention_norm/vars": [
145
+ "model_00000.weights.h5"
146
+ ],
147
+ "/layers/gemma_decoder_layer_5/mlp/down/vars": [
148
+ "model_00000.weights.h5"
149
+ ],
150
+ "/layers/gemma_decoder_layer_5/mlp/gate/vars": [
151
+ "model_00000.weights.h5"
152
+ ],
153
+ "/layers/gemma_decoder_layer_5/mlp/up/vars": [
154
+ "model_00000.weights.h5"
155
+ ],
156
+ "/layers/gemma_decoder_layer_5/mlp_norm/vars": [
157
+ "model_00000.weights.h5"
158
+ ],
159
+ "/layers/gemma_decoder_layer_5/attention/key/vars": [
160
+ "model_00000.weights.h5"
161
+ ],
162
+ "/layers/gemma_decoder_layer_5/attention/output_proj/vars": [
163
+ "model_00000.weights.h5"
164
+ ],
165
+ "/layers/gemma_decoder_layer_5/attention/query/vars": [
166
+ "model_00000.weights.h5"
167
+ ],
168
+ "/layers/gemma_decoder_layer_5/attention/value/vars": [
169
+ "model_00000.weights.h5"
170
+ ],
171
+ "/layers/gemma_decoder_layer_6/attention/key/vars": [
172
+ "model_00000.weights.h5"
173
+ ],
174
+ "/layers/gemma_decoder_layer_6/attention/output_proj/vars": [
175
+ "model_00000.weights.h5"
176
+ ],
177
+ "/layers/gemma_decoder_layer_6/attention/query/vars": [
178
+ "model_00000.weights.h5"
179
+ ],
180
+ "/layers/gemma_decoder_layer_6/attention/value/vars": [
181
+ "model_00000.weights.h5"
182
+ ],
183
+ "/layers/gemma_decoder_layer_10/attention_norm/vars": [
184
+ "model_00000.weights.h5"
185
+ ],
186
+ "/layers/gemma_decoder_layer_10/mlp/down/vars": [
187
+ "model_00000.weights.h5"
188
+ ],
189
+ "/layers/gemma_decoder_layer_10/mlp/gate/vars": [
190
+ "model_00000.weights.h5"
191
+ ],
192
+ "/layers/gemma_decoder_layer_10/mlp/up/vars": [
193
+ "model_00001.weights.h5"
194
+ ],
195
+ "/layers/gemma_decoder_layer_10/mlp_norm/vars": [
196
+ "model_00001.weights.h5"
197
+ ],
198
+ "/layers/gemma_decoder_layer_10/attention/key/vars": [
199
+ "model_00001.weights.h5"
200
+ ],
201
+ "/layers/gemma_decoder_layer_10/attention/output_proj/vars": [
202
+ "model_00001.weights.h5"
203
+ ],
204
+ "/layers/gemma_decoder_layer_10/attention/query/vars": [
205
+ "model_00001.weights.h5"
206
+ ],
207
+ "/layers/gemma_decoder_layer_10/attention/value/vars": [
208
+ "model_00001.weights.h5"
209
+ ],
210
+ "/layers/gemma_decoder_layer_11/attention_norm/vars": [
211
+ "model_00001.weights.h5"
212
+ ],
213
+ "/layers/gemma_decoder_layer_11/mlp/down/vars": [
214
+ "model_00001.weights.h5"
215
+ ],
216
+ "/layers/gemma_decoder_layer_11/mlp/gate/vars": [
217
+ "model_00001.weights.h5"
218
+ ],
219
+ "/layers/gemma_decoder_layer_11/mlp/up/vars": [
220
+ "model_00001.weights.h5"
221
+ ],
222
+ "/layers/gemma_decoder_layer_11/mlp_norm/vars": [
223
+ "model_00001.weights.h5"
224
+ ],
225
+ "/layers/gemma_decoder_layer_11/attention/key/vars": [
226
+ "model_00001.weights.h5"
227
+ ],
228
+ "/layers/gemma_decoder_layer_11/attention/output_proj/vars": [
229
+ "model_00001.weights.h5"
230
+ ],
231
+ "/layers/gemma_decoder_layer_11/attention/query/vars": [
232
+ "model_00001.weights.h5"
233
+ ],
234
+ "/layers/gemma_decoder_layer_11/attention/value/vars": [
235
+ "model_00001.weights.h5"
236
+ ],
237
+ "/layers/gemma_decoder_layer_12/attention_norm/vars": [
238
+ "model_00001.weights.h5"
239
+ ],
240
+ "/layers/gemma_decoder_layer_12/mlp/down/vars": [
241
+ "model_00001.weights.h5"
242
+ ],
243
+ "/layers/gemma_decoder_layer_12/mlp/gate/vars": [
244
+ "model_00001.weights.h5"
245
+ ],
246
+ "/layers/gemma_decoder_layer_12/mlp/up/vars": [
247
+ "model_00001.weights.h5"
248
+ ],
249
+ "/layers/gemma_decoder_layer_12/mlp_norm/vars": [
250
+ "model_00001.weights.h5"
251
+ ],
252
+ "/layers/gemma_decoder_layer_12/attention/key/vars": [
253
+ "model_00001.weights.h5"
254
+ ],
255
+ "/layers/gemma_decoder_layer_12/attention/output_proj/vars": [
256
+ "model_00001.weights.h5"
257
+ ],
258
+ "/layers/gemma_decoder_layer_12/attention/query/vars": [
259
+ "model_00001.weights.h5"
260
+ ],
261
+ "/layers/gemma_decoder_layer_12/attention/value/vars": [
262
+ "model_00001.weights.h5"
263
+ ],
264
+ "/layers/gemma_decoder_layer_13/attention_norm/vars": [
265
+ "model_00001.weights.h5"
266
+ ],
267
+ "/layers/gemma_decoder_layer_13/mlp/down/vars": [
268
+ "model_00001.weights.h5"
269
+ ],
270
+ "/layers/gemma_decoder_layer_13/mlp/gate/vars": [
271
+ "model_00001.weights.h5"
272
+ ],
273
+ "/layers/gemma_decoder_layer_13/mlp/up/vars": [
274
+ "model_00001.weights.h5"
275
+ ],
276
+ "/layers/gemma_decoder_layer_13/mlp_norm/vars": [
277
+ "model_00001.weights.h5"
278
+ ],
279
+ "/layers/gemma_decoder_layer_13/attention/key/vars": [
280
+ "model_00001.weights.h5"
281
+ ],
282
+ "/layers/gemma_decoder_layer_13/attention/output_proj/vars": [
283
+ "model_00001.weights.h5"
284
+ ],
285
+ "/layers/gemma_decoder_layer_13/attention/query/vars": [
286
+ "model_00001.weights.h5"
287
+ ],
288
+ "/layers/gemma_decoder_layer_13/attention/value/vars": [
289
+ "model_00001.weights.h5"
290
+ ],
291
+ "/layers/gemma_decoder_layer_14/attention_norm/vars": [
292
+ "model_00001.weights.h5"
293
+ ],
294
+ "/layers/gemma_decoder_layer_14/mlp/down/vars": [
295
+ "model_00001.weights.h5"
296
+ ],
297
+ "/layers/gemma_decoder_layer_14/mlp/gate/vars": [
298
+ "model_00001.weights.h5"
299
+ ],
300
+ "/layers/gemma_decoder_layer_14/mlp/up/vars": [
301
+ "model_00001.weights.h5"
302
+ ],
303
+ "/layers/gemma_decoder_layer_14/mlp_norm/vars": [
304
+ "model_00001.weights.h5"
305
+ ],
306
+ "/layers/gemma_decoder_layer_14/attention/key/vars": [
307
+ "model_00001.weights.h5"
308
+ ],
309
+ "/layers/gemma_decoder_layer_14/attention/output_proj/vars": [
310
+ "model_00001.weights.h5"
311
+ ],
312
+ "/layers/gemma_decoder_layer_14/attention/query/vars": [
313
+ "model_00001.weights.h5"
314
+ ],
315
+ "/layers/gemma_decoder_layer_14/attention/value/vars": [
316
+ "model_00001.weights.h5"
317
+ ],
318
+ "/layers/gemma_decoder_layer_15/attention/key/vars": [
319
+ "model_00001.weights.h5"
320
+ ],
321
+ "/layers/gemma_decoder_layer_15/attention/output_proj/vars": [
322
+ "model_00001.weights.h5"
323
+ ],
324
+ "/layers/gemma_decoder_layer_15/attention/query/vars": [
325
+ "model_00001.weights.h5"
326
+ ],
327
+ "/layers/gemma_decoder_layer_15/attention/value/vars": [
328
+ "model_00001.weights.h5"
329
+ ],
330
+ "/layers/gemma_decoder_layer_6/attention_norm/vars": [
331
+ "model_00001.weights.h5"
332
+ ],
333
+ "/layers/gemma_decoder_layer_6/mlp/down/vars": [
334
+ "model_00001.weights.h5"
335
+ ],
336
+ "/layers/gemma_decoder_layer_6/mlp/gate/vars": [
337
+ "model_00001.weights.h5"
338
+ ],
339
+ "/layers/gemma_decoder_layer_6/mlp/up/vars": [
340
+ "model_00001.weights.h5"
341
+ ],
342
+ "/layers/gemma_decoder_layer_6/mlp_norm/vars": [
343
+ "model_00001.weights.h5"
344
+ ],
345
+ "/layers/gemma_decoder_layer_7/attention_norm/vars": [
346
+ "model_00001.weights.h5"
347
+ ],
348
+ "/layers/gemma_decoder_layer_7/mlp/down/vars": [
349
+ "model_00001.weights.h5"
350
+ ],
351
+ "/layers/gemma_decoder_layer_7/mlp/gate/vars": [
352
+ "model_00001.weights.h5"
353
+ ],
354
+ "/layers/gemma_decoder_layer_7/mlp/up/vars": [
355
+ "model_00001.weights.h5"
356
+ ],
357
+ "/layers/gemma_decoder_layer_7/mlp_norm/vars": [
358
+ "model_00001.weights.h5"
359
+ ],
360
+ "/layers/gemma_decoder_layer_7/attention/key/vars": [
361
+ "model_00001.weights.h5"
362
+ ],
363
+ "/layers/gemma_decoder_layer_7/attention/output_proj/vars": [
364
+ "model_00001.weights.h5"
365
+ ],
366
+ "/layers/gemma_decoder_layer_7/attention/query/vars": [
367
+ "model_00001.weights.h5"
368
+ ],
369
+ "/layers/gemma_decoder_layer_7/attention/value/vars": [
370
+ "model_00001.weights.h5"
371
+ ],
372
+ "/layers/gemma_decoder_layer_8/attention_norm/vars": [
373
+ "model_00001.weights.h5"
374
+ ],
375
+ "/layers/gemma_decoder_layer_8/mlp/down/vars": [
376
+ "model_00001.weights.h5"
377
+ ],
378
+ "/layers/gemma_decoder_layer_8/mlp/gate/vars": [
379
+ "model_00001.weights.h5"
380
+ ],
381
+ "/layers/gemma_decoder_layer_8/mlp/up/vars": [
382
+ "model_00001.weights.h5"
383
+ ],
384
+ "/layers/gemma_decoder_layer_8/mlp_norm/vars": [
385
+ "model_00001.weights.h5"
386
+ ],
387
+ "/layers/gemma_decoder_layer_8/attention/key/vars": [
388
+ "model_00001.weights.h5"
389
+ ],
390
+ "/layers/gemma_decoder_layer_8/attention/output_proj/vars": [
391
+ "model_00001.weights.h5"
392
+ ],
393
+ "/layers/gemma_decoder_layer_8/attention/query/vars": [
394
+ "model_00001.weights.h5"
395
+ ],
396
+ "/layers/gemma_decoder_layer_8/attention/value/vars": [
397
+ "model_00001.weights.h5"
398
+ ],
399
+ "/layers/gemma_decoder_layer_9/attention_norm/vars": [
400
+ "model_00001.weights.h5"
401
+ ],
402
+ "/layers/gemma_decoder_layer_9/mlp/down/vars": [
403
+ "model_00001.weights.h5"
404
+ ],
405
+ "/layers/gemma_decoder_layer_9/mlp/gate/vars": [
406
+ "model_00001.weights.h5"
407
+ ],
408
+ "/layers/gemma_decoder_layer_9/mlp/up/vars": [
409
+ "model_00001.weights.h5"
410
+ ],
411
+ "/layers/gemma_decoder_layer_9/mlp_norm/vars": [
412
+ "model_00001.weights.h5"
413
+ ],
414
+ "/layers/gemma_decoder_layer_9/attention/key/vars": [
415
+ "model_00001.weights.h5"
416
+ ],
417
+ "/layers/gemma_decoder_layer_9/attention/output_proj/vars": [
418
+ "model_00001.weights.h5"
419
+ ],
420
+ "/layers/gemma_decoder_layer_9/attention/query/vars": [
421
+ "model_00001.weights.h5"
422
+ ],
423
+ "/layers/gemma_decoder_layer_9/attention/value/vars": [
424
+ "model_00001.weights.h5"
425
+ ],
426
+ "/layers/gemma_decoder_layer_15/attention_norm/vars": [
427
+ "model_00001.weights.h5"
428
+ ],
429
+ "/layers/gemma_decoder_layer_15/mlp/down/vars": [
430
+ "model_00001.weights.h5"
431
+ ],
432
+ "/layers/gemma_decoder_layer_15/mlp/gate/vars": [
433
+ "model_00001.weights.h5"
434
+ ],
435
+ "/layers/gemma_decoder_layer_15/mlp/up/vars": [
436
+ "model_00001.weights.h5"
437
+ ],
438
+ "/layers/gemma_decoder_layer_15/mlp_norm/vars": [
439
+ "model_00001.weights.h5"
440
+ ],
441
+ "/layers/gemma_decoder_layer_16/attention_norm/vars": [
442
+ "model_00001.weights.h5"
443
+ ],
444
+ "/layers/gemma_decoder_layer_16/mlp/down/vars": [
445
+ "model_00001.weights.h5"
446
+ ],
447
+ "/layers/gemma_decoder_layer_16/mlp/gate/vars": [
448
+ "model_00002.weights.h5"
449
+ ],
450
+ "/layers/gemma_decoder_layer_16/mlp/up/vars": [
451
+ "model_00002.weights.h5"
452
+ ],
453
+ "/layers/gemma_decoder_layer_16/mlp_norm/vars": [
454
+ "model_00002.weights.h5"
455
+ ],
456
+ "/layers/gemma_decoder_layer_16/attention/key/vars": [
457
+ "model_00002.weights.h5"
458
+ ],
459
+ "/layers/gemma_decoder_layer_16/attention/output_proj/vars": [
460
+ "model_00002.weights.h5"
461
+ ],
462
+ "/layers/gemma_decoder_layer_16/attention/query/vars": [
463
+ "model_00002.weights.h5"
464
+ ],
465
+ "/layers/gemma_decoder_layer_16/attention/value/vars": [
466
+ "model_00002.weights.h5"
467
+ ],
468
+ "/layers/gemma_decoder_layer_17/attention_norm/vars": [
469
+ "model_00002.weights.h5"
470
+ ],
471
+ "/layers/gemma_decoder_layer_17/mlp/down/vars": [
472
+ "model_00002.weights.h5"
473
+ ],
474
+ "/layers/gemma_decoder_layer_17/mlp/gate/vars": [
475
+ "model_00002.weights.h5"
476
+ ],
477
+ "/layers/gemma_decoder_layer_17/mlp/up/vars": [
478
+ "model_00002.weights.h5"
479
+ ],
480
+ "/layers/gemma_decoder_layer_17/mlp_norm/vars": [
481
+ "model_00002.weights.h5"
482
+ ],
483
+ "/layers/gemma_decoder_layer_17/attention/key/vars": [
484
+ "model_00002.weights.h5"
485
+ ],
486
+ "/layers/gemma_decoder_layer_17/attention/output_proj/vars": [
487
+ "model_00002.weights.h5"
488
+ ],
489
+ "/layers/gemma_decoder_layer_17/attention/query/vars": [
490
+ "model_00002.weights.h5"
491
+ ],
492
+ "/layers/gemma_decoder_layer_17/attention/value/vars": [
493
+ "model_00002.weights.h5"
494
+ ],
495
+ "/layers/gemma_decoder_layer_18/attention_norm/vars": [
496
+ "model_00002.weights.h5"
497
+ ],
498
+ "/layers/gemma_decoder_layer_18/mlp/down/vars": [
499
+ "model_00002.weights.h5"
500
+ ],
501
+ "/layers/gemma_decoder_layer_18/mlp/gate/vars": [
502
+ "model_00002.weights.h5"
503
+ ],
504
+ "/layers/gemma_decoder_layer_18/mlp/up/vars": [
505
+ "model_00002.weights.h5"
506
+ ],
507
+ "/layers/gemma_decoder_layer_18/mlp_norm/vars": [
508
+ "model_00002.weights.h5"
509
+ ],
510
+ "/layers/gemma_decoder_layer_18/attention/key/vars": [
511
+ "model_00002.weights.h5"
512
+ ],
513
+ "/layers/gemma_decoder_layer_18/attention/output_proj/vars": [
514
+ "model_00002.weights.h5"
515
+ ],
516
+ "/layers/gemma_decoder_layer_18/attention/query/vars": [
517
+ "model_00002.weights.h5"
518
+ ],
519
+ "/layers/gemma_decoder_layer_18/attention/value/vars": [
520
+ "model_00002.weights.h5"
521
+ ],
522
+ "/layers/gemma_decoder_layer_19/attention_norm/vars": [
523
+ "model_00002.weights.h5"
524
+ ],
525
+ "/layers/gemma_decoder_layer_19/mlp/down/vars": [
526
+ "model_00002.weights.h5"
527
+ ],
528
+ "/layers/gemma_decoder_layer_19/mlp/gate/vars": [
529
+ "model_00002.weights.h5"
530
+ ],
531
+ "/layers/gemma_decoder_layer_19/mlp/up/vars": [
532
+ "model_00002.weights.h5"
533
+ ],
534
+ "/layers/gemma_decoder_layer_19/mlp_norm/vars": [
535
+ "model_00002.weights.h5"
536
+ ],
537
+ "/layers/gemma_decoder_layer_19/attention/key/vars": [
538
+ "model_00002.weights.h5"
539
+ ],
540
+ "/layers/gemma_decoder_layer_19/attention/output_proj/vars": [
541
+ "model_00002.weights.h5"
542
+ ],
543
+ "/layers/gemma_decoder_layer_19/attention/query/vars": [
544
+ "model_00002.weights.h5"
545
+ ],
546
+ "/layers/gemma_decoder_layer_19/attention/value/vars": [
547
+ "model_00002.weights.h5"
548
+ ],
549
+ "/layers/gemma_decoder_layer_20/attention_norm/vars": [
550
+ "model_00002.weights.h5"
551
+ ],
552
+ "/layers/gemma_decoder_layer_20/mlp/down/vars": [
553
+ "model_00002.weights.h5"
554
+ ],
555
+ "/layers/gemma_decoder_layer_20/mlp/gate/vars": [
556
+ "model_00002.weights.h5"
557
+ ],
558
+ "/layers/gemma_decoder_layer_20/mlp/up/vars": [
559
+ "model_00002.weights.h5"
560
+ ],
561
+ "/layers/gemma_decoder_layer_20/mlp_norm/vars": [
562
+ "model_00002.weights.h5"
563
+ ],
564
+ "/layers/gemma_decoder_layer_20/attention/key/vars": [
565
+ "model_00002.weights.h5"
566
+ ],
567
+ "/layers/gemma_decoder_layer_20/attention/output_proj/vars": [
568
+ "model_00002.weights.h5"
569
+ ],
570
+ "/layers/gemma_decoder_layer_20/attention/query/vars": [
571
+ "model_00002.weights.h5"
572
+ ],
573
+ "/layers/gemma_decoder_layer_20/attention/value/vars": [
574
+ "model_00002.weights.h5"
575
+ ],
576
+ "/layers/gemma_decoder_layer_21/attention_norm/vars": [
577
+ "model_00002.weights.h5"
578
+ ],
579
+ "/layers/gemma_decoder_layer_21/mlp/down/vars": [
580
+ "model_00002.weights.h5"
581
+ ],
582
+ "/layers/gemma_decoder_layer_21/mlp/gate/vars": [
583
+ "model_00002.weights.h5"
584
+ ],
585
+ "/layers/gemma_decoder_layer_21/mlp/up/vars": [
586
+ "model_00002.weights.h5"
587
+ ],
588
+ "/layers/gemma_decoder_layer_21/mlp_norm/vars": [
589
+ "model_00002.weights.h5"
590
+ ],
591
+ "/layers/gemma_decoder_layer_21/attention/key/vars": [
592
+ "model_00002.weights.h5"
593
+ ],
594
+ "/layers/gemma_decoder_layer_21/attention/output_proj/vars": [
595
+ "model_00002.weights.h5"
596
+ ],
597
+ "/layers/gemma_decoder_layer_21/attention/query/vars": [
598
+ "model_00002.weights.h5"
599
+ ],
600
+ "/layers/gemma_decoder_layer_21/attention/value/vars": [
601
+ "model_00002.weights.h5"
602
+ ],
603
+ "/layers/gemma_decoder_layer_22/attention_norm/vars": [
604
+ "model_00002.weights.h5"
605
+ ],
606
+ "/layers/gemma_decoder_layer_22/mlp/down/vars": [
607
+ "model_00002.weights.h5"
608
+ ],
609
+ "/layers/gemma_decoder_layer_22/mlp/gate/vars": [
610
+ "model_00002.weights.h5"
611
+ ],
612
+ "/layers/gemma_decoder_layer_22/mlp/up/vars": [
613
+ "model_00002.weights.h5"
614
+ ],
615
+ "/layers/gemma_decoder_layer_22/mlp_norm/vars": [
616
+ "model_00002.weights.h5"
617
+ ],
618
+ "/layers/gemma_decoder_layer_22/attention/key/vars": [
619
+ "model_00002.weights.h5"
620
+ ],
621
+ "/layers/gemma_decoder_layer_22/attention/output_proj/vars": [
622
+ "model_00002.weights.h5"
623
+ ],
624
+ "/layers/gemma_decoder_layer_22/attention/query/vars": [
625
+ "model_00002.weights.h5"
626
+ ],
627
+ "/layers/gemma_decoder_layer_22/attention/value/vars": [
628
+ "model_00002.weights.h5"
629
+ ],
630
+ "/layers/gemma_decoder_layer_23/attention_norm/vars": [
631
+ "model_00002.weights.h5"
632
+ ],
633
+ "/layers/gemma_decoder_layer_23/mlp/down/vars": [
634
+ "model_00002.weights.h5"
635
+ ],
636
+ "/layers/gemma_decoder_layer_23/mlp/gate/vars": [
637
+ "model_00002.weights.h5"
638
+ ],
639
+ "/layers/gemma_decoder_layer_23/mlp/up/vars": [
640
+ "model_00002.weights.h5"
641
+ ],
642
+ "/layers/gemma_decoder_layer_23/mlp_norm/vars": [
643
+ "model_00002.weights.h5"
644
+ ],
645
+ "/layers/gemma_decoder_layer_23/attention/key/vars": [
646
+ "model_00002.weights.h5"
647
+ ],
648
+ "/layers/gemma_decoder_layer_23/attention/output_proj/vars": [
649
+ "model_00002.weights.h5"
650
+ ],
651
+ "/layers/gemma_decoder_layer_23/attention/query/vars": [
652
+ "model_00002.weights.h5"
653
+ ],
654
+ "/layers/gemma_decoder_layer_23/attention/value/vars": [
655
+ "model_00002.weights.h5"
656
+ ],
657
+ "/layers/gemma_decoder_layer_24/attention/key/vars": [
658
+ "model_00002.weights.h5"
659
+ ],
660
+ "/layers/gemma_decoder_layer_24/attention/output_proj/vars": [
661
+ "model_00002.weights.h5"
662
+ ],
663
+ "/layers/gemma_decoder_layer_24/attention/query/vars": [
664
+ "model_00002.weights.h5"
665
+ ],
666
+ "/layers/gemma_decoder_layer_24/attention/value/vars": [
667
+ "model_00002.weights.h5"
668
+ ],
669
+ "/layers/gemma_decoder_layer_24/attention_norm/vars": [
670
+ "model_00002.weights.h5"
671
+ ],
672
+ "/layers/gemma_decoder_layer_24/mlp/down/vars": [
673
+ "model_00002.weights.h5"
674
+ ],
675
+ "/layers/gemma_decoder_layer_24/mlp/gate/vars": [
676
+ "model_00002.weights.h5"
677
+ ],
678
+ "/layers/gemma_decoder_layer_24/mlp/up/vars": [
679
+ "model_00002.weights.h5"
680
+ ],
681
+ "/layers/gemma_decoder_layer_24/mlp_norm/vars": [
682
+ "model_00002.weights.h5"
683
+ ],
684
+ "/layers/gemma_decoder_layer_25/attention_norm/vars": [
685
+ "model_00002.weights.h5"
686
+ ],
687
+ "/layers/gemma_decoder_layer_25/mlp/down/vars": [
688
+ "model_00002.weights.h5"
689
+ ],
690
+ "/layers/gemma_decoder_layer_25/mlp/gate/vars": [
691
+ "model_00002.weights.h5"
692
+ ],
693
+ "/layers/gemma_decoder_layer_25/mlp/up/vars": [
694
+ "model_00002.weights.h5"
695
+ ],
696
+ "/layers/gemma_decoder_layer_25/mlp_norm/vars": [
697
+ "model_00002.weights.h5"
698
+ ],
699
+ "/layers/gemma_decoder_layer_25/attention/key/vars": [
700
+ "model_00002.weights.h5"
701
+ ],
702
+ "/layers/gemma_decoder_layer_25/attention/output_proj/vars": [
703
+ "model_00002.weights.h5"
704
+ ],
705
+ "/layers/gemma_decoder_layer_25/attention/query/vars": [
706
+ "model_00002.weights.h5"
707
+ ],
708
+ "/layers/gemma_decoder_layer_25/attention/value/vars": [
709
+ "model_00003.weights.h5"
710
+ ],
711
+ "/layers/gemma_decoder_layer_26/attention_norm/vars": [
712
+ "model_00003.weights.h5"
713
+ ],
714
+ "/layers/gemma_decoder_layer_26/mlp/down/vars": [
715
+ "model_00003.weights.h5"
716
+ ],
717
+ "/layers/gemma_decoder_layer_26/mlp/gate/vars": [
718
+ "model_00003.weights.h5"
719
+ ],
720
+ "/layers/gemma_decoder_layer_26/mlp/up/vars": [
721
+ "model_00003.weights.h5"
722
+ ],
723
+ "/layers/gemma_decoder_layer_26/mlp_norm/vars": [
724
+ "model_00003.weights.h5"
725
+ ],
726
+ "/layers/gemma_decoder_layer_26/attention/key/vars": [
727
+ "model_00003.weights.h5"
728
+ ],
729
+ "/layers/gemma_decoder_layer_26/attention/output_proj/vars": [
730
+ "model_00003.weights.h5"
731
+ ],
732
+ "/layers/gemma_decoder_layer_26/attention/query/vars": [
733
+ "model_00003.weights.h5"
734
+ ],
735
+ "/layers/gemma_decoder_layer_26/attention/value/vars": [
736
+ "model_00003.weights.h5"
737
+ ],
738
+ "/layers/gemma_decoder_layer_27/attention_norm/vars": [
739
+ "model_00003.weights.h5"
740
+ ],
741
+ "/layers/gemma_decoder_layer_27/mlp/down/vars": [
742
+ "model_00003.weights.h5"
743
+ ],
744
+ "/layers/gemma_decoder_layer_27/mlp/gate/vars": [
745
+ "model_00003.weights.h5"
746
+ ],
747
+ "/layers/gemma_decoder_layer_27/mlp/up/vars": [
748
+ "model_00003.weights.h5"
749
+ ],
750
+ "/layers/gemma_decoder_layer_27/mlp_norm/vars": [
751
+ "model_00003.weights.h5"
752
+ ],
753
+ "/layers/gemma_decoder_layer_27/attention/key/vars": [
754
+ "model_00003.weights.h5"
755
+ ],
756
+ "/layers/gemma_decoder_layer_27/attention/output_proj/vars": [
757
+ "model_00003.weights.h5"
758
+ ],
759
+ "/layers/gemma_decoder_layer_27/attention/query/vars": [
760
+ "model_00003.weights.h5"
761
+ ],
762
+ "/layers/gemma_decoder_layer_27/attention/value/vars": [
763
+ "model_00003.weights.h5"
764
+ ],
765
+ "/layers/gemma_rms_norm/vars": [
766
+ "model_00003.weights.h5"
767
+ ]
768
+ }
769
+ }
model_00000.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9b8d090c46139dde9b984240821be830ff8ee5ca9b7ab7d7b0ed76038d4bdab
3
+ size 5299378752
model_00001.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dd87c57a78fee58524bbe833dac987e0a76ded986b40b5c3bcd55120367c4c6
3
+ size 5286855232
model_00002.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dec4d6b1e8b96d30335ad9211cd6209b1f0fbd89fecf26b672d0b6f11689df7c
3
+ size 5362346560
model_00003.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7a5c347ebcd90ee321d0de3acde6599ceb1f92d03b9f7b59ad6e55d1c71e62b
3
+ size 1134357056
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c15eb04bc5ad609fb26533e8525302c5640a945e5f67f65b7c849900acda7d99
3
+ size 17518497