schroneko commited on
Commit
33b84b0
·
verified ·
1 Parent(s): 5cc769c

Add files using upload-large-folder tool

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -7,15 +7,15 @@ extra_gated_prompt: To access Gemma on Hugging Face, you’re required to review
7
  agree to Google’s usage license. To do this, please ensure you’re logged in to Hugging
8
  Face and click below. Requests are processed immediately.
9
  extra_gated_button_content: Acknowledge license
10
- base_model: google/translategemma-27b-it
11
  tags:
12
  - mlx
 
13
  ---
14
 
15
- # mlx-community/translategemma-27b-it-8bit
16
 
17
- This model [mlx-community/translategemma-27b-it-8bit](https://huggingface.co/mlx-community/translategemma-27b-it-8bit) was
18
- converted to MLX format from [google/translategemma-27b-it](https://huggingface.co/google/translategemma-27b-it)
19
  using mlx-lm version **0.29.1**.
20
 
21
  ## Use with mlx
@@ -27,7 +27,7 @@ pip install mlx-lm
27
  ```python
28
  from mlx_lm import load, generate
29
 
30
- model, tokenizer = load("mlx-community/translategemma-27b-it-8bit")
31
 
32
  prompt = "hello"
33
 
 
7
  agree to Google’s usage license. To do this, please ensure you’re logged in to Hugging
8
  Face and click below. Requests are processed immediately.
9
  extra_gated_button_content: Acknowledge license
 
10
  tags:
11
  - mlx
12
+ base_model: google/translategemma-12b-it
13
  ---
14
 
15
+ # mlx-community/translategemma-12b-it-8bit
16
 
17
+ This model [mlx-community/translategemma-12b-it-8bit](https://huggingface.co/mlx-community/translategemma-12b-it-8bit) was
18
+ converted to MLX format from [google/translategemma-12b-it](https://huggingface.co/google/translategemma-12b-it)
19
  using mlx-lm version **0.29.1**.
20
 
21
  ## Use with mlx
 
27
  ```python
28
  from mlx_lm import load, generate
29
 
30
+ model, tokenizer = load("mlx-community/translategemma-12b-it-8bit")
31
 
32
  prompt = "hello"
33
 
config.json CHANGED
@@ -26,11 +26,11 @@
26
  "attn_logit_softcapping": null,
27
  "dtype": "bfloat16",
28
  "final_logit_softcapping": null,
29
- "head_dim": 128,
30
  "hidden_activation": "gelu_pytorch_tanh",
31
- "hidden_size": 5376,
32
  "initializer_range": 0.02,
33
- "intermediate_size": 21504,
34
  "layer_types": [
35
  "sliding_attention",
36
  "sliding_attention",
@@ -79,28 +79,14 @@
79
  "sliding_attention",
80
  "sliding_attention",
81
  "sliding_attention",
82
- "full_attention",
83
- "sliding_attention",
84
- "sliding_attention",
85
- "sliding_attention",
86
- "sliding_attention",
87
- "sliding_attention",
88
- "full_attention",
89
- "sliding_attention",
90
- "sliding_attention",
91
- "sliding_attention",
92
- "sliding_attention",
93
- "sliding_attention",
94
- "full_attention",
95
- "sliding_attention",
96
- "sliding_attention"
97
  ],
98
  "max_position_embeddings": 131072,
99
  "model_type": "gemma3_text",
100
- "num_attention_heads": 32,
101
- "num_hidden_layers": 62,
102
- "num_key_value_heads": 16,
103
- "query_pre_attn_scalar": 168,
104
  "rms_norm_eps": 1e-06,
105
  "rope_local_base_freq": 10000,
106
  "rope_parameters": {
 
26
  "attn_logit_softcapping": null,
27
  "dtype": "bfloat16",
28
  "final_logit_softcapping": null,
29
+ "head_dim": 256,
30
  "hidden_activation": "gelu_pytorch_tanh",
31
+ "hidden_size": 3840,
32
  "initializer_range": 0.02,
33
+ "intermediate_size": 15360,
34
  "layer_types": [
35
  "sliding_attention",
36
  "sliding_attention",
 
79
  "sliding_attention",
80
  "sliding_attention",
81
  "sliding_attention",
82
+ "full_attention"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
83
  ],
84
  "max_position_embeddings": 131072,
85
  "model_type": "gemma3_text",
86
+ "num_attention_heads": 16,
87
+ "num_hidden_layers": 48,
88
+ "num_key_value_heads": 8,
89
+ "query_pre_attn_scalar": 256,
90
  "rms_norm_eps": 1e-06,
91
  "rope_local_base_freq": 10000,
92
  "rope_parameters": {
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d375791d8b385012539bae9407566686e14b0009b106bfaf9997bc85f4b9069
3
+ size 5356989205
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28d257759581335491b1a32ec426a08166e63f773e52ad7b82535d8ffdda157b
3
+ size 5352694354
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:314fd90fe022d1e6a6fd556eb0542bfa66b82fea2437a62cbc4ab078ee7cde3c
3
+ size 1792614154
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff