Hugging Face's logo

rudrashah
/
RLM_hingu

Text Generation
KerasHub
Hindi
English
hinglish
Model card Files
xet
Community
rudrashah commited on
Commit
9c461fe
·
verified ·
1 Parent(s): e469e30

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. assets/tokenizer/vocabulary.spm +3 -0
  3. config.json +30 -0
  4. metadata.json +9 -0
  5. model.weights.h5 +3 -0
  6. preprocessor.json +45 -0
  7. task.json +83 -0
  8. tokenizer.json +22 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ assets/tokenizer/vocabulary.spm filter=lfs diff=lfs merge=lfs -text
assets/tokenizer/vocabulary.spm ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "module": "keras_hub.src.models.gemma3.gemma3_backbone",
3
+ "class_name": "Gemma3Backbone",
4
+ "config": {
5
+ "name": "gemma3_backbone",
6
+ "trainable": true,
7
+ "vocabulary_size": 262144,
8
+ "image_size": null,
9
+ "num_layers": 26,
10
+ "num_query_heads": 4,
11
+ "num_key_value_heads": 1,
12
+ "hidden_dim": 1152,
13
+ "intermediate_dim": 6912,
14
+ "head_dim": 256,
15
+ "query_head_dim_normalize": true,
16
+ "use_query_key_norm": true,
17
+ "use_post_ffw_norm": true,
18
+ "use_post_attention_norm": true,
19
+ "attention_logit_soft_cap": null,
20
+ "final_logit_soft_cap": null,
21
+ "use_sliding_window_attention": true,
22
+ "sliding_window_size": 512,
23
+ "local_rope_scaling_factor": 1.0,
24
+ "global_rope_scaling_factor": 1.0,
25
+ "vision_encoder": null,
26
+ "layer_norm_epsilon": 1e-06,
27
+ "dropout": 0
28
+ },
29
+ "registered_name": "keras_hub>Gemma3Backbone"
30
+ }
metadata.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "keras_version": "3.9.2",
3
+ "keras_hub_version": "0.20.0",
4
+ "parameter_count": 1010322560,
5
+ "date_saved": "2025-05-05@15:28:22",
6
+ "tasks": [
7
+ "CausalLM"
8
+ ]
9
+ }
model.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f2b0bb196310bab456e6cc7a2d32d2df39d392aa8d5e47b275b7151224f5045
3
+ size 4000656984
preprocessor.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "module": "keras_hub.src.models.gemma3.gemma3_causal_lm_preprocessor",
3
+ "class_name": "Gemma3CausalLMPreprocessor",
4
+ "config": {
5
+ "name": "gemma3_causal_lm_preprocessor",
6
+ "trainable": true,
7
+ "dtype": {
8
+ "module": "keras",
9
+ "class_name": "DTypePolicy",
10
+ "config": {
11
+ "name": "float32"
12
+ },
13
+ "registered_name": null
14
+ },
15
+ "tokenizer": {
16
+ "module": "keras_hub.src.models.gemma3.gemma3_tokenizer",
17
+ "class_name": "Gemma3Tokenizer",
18
+ "config": {
19
+ "name": "gemma3_tokenizer",
20
+ "trainable": true,
21
+ "dtype": {
22
+ "module": "keras",
23
+ "class_name": "DTypePolicy",
24
+ "config": {
25
+ "name": "int32"
26
+ },
27
+ "registered_name": null
28
+ },
29
+ "config_file": "tokenizer.json",
30
+ "proto": null,
31
+ "sequence_length": null,
32
+ "add_bos": false,
33
+ "add_eos": false
34
+ },
35
+ "registered_name": "keras_hub>Gemma3Tokenizer"
36
+ },
37
+ "config_file": "preprocessor.json",
38
+ "sequence_length": 256,
39
+ "add_start_token": true,
40
+ "add_end_token": true,
41
+ "num_vision_tokens_per_image": 0,
42
+ "max_images_per_prompt": 0
43
+ },
44
+ "registered_name": "keras_hub>Gemma3CausalLMPreprocessor"
45
+ }
task.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "module": "keras_hub.src.models.gemma3.gemma3_causal_lm",
3
+ "class_name": "Gemma3CausalLM",
4
+ "config": {
5
+ "backbone": {
6
+ "module": "keras_hub.src.models.gemma3.gemma3_backbone",
7
+ "class_name": "Gemma3Backbone",
8
+ "config": {
9
+ "name": "gemma3_backbone",
10
+ "trainable": true,
11
+ "vocabulary_size": 262144,
12
+ "image_size": null,
13
+ "num_layers": 26,
14
+ "num_query_heads": 4,
15
+ "num_key_value_heads": 1,
16
+ "hidden_dim": 1152,
17
+ "intermediate_dim": 6912,
18
+ "head_dim": 256,
19
+ "query_head_dim_normalize": true,
20
+ "use_query_key_norm": true,
21
+ "use_post_ffw_norm": true,
22
+ "use_post_attention_norm": true,
23
+ "attention_logit_soft_cap": null,
24
+ "final_logit_soft_cap": null,
25
+ "use_sliding_window_attention": true,
26
+ "sliding_window_size": 512,
27
+ "local_rope_scaling_factor": 1.0,
28
+ "global_rope_scaling_factor": 1.0,
29
+ "vision_encoder": null,
30
+ "layer_norm_epsilon": 1e-06,
31
+ "dropout": 0
32
+ },
33
+ "registered_name": "keras_hub>Gemma3Backbone"
34
+ },
35
+ "preprocessor": {
36
+ "module": "keras_hub.src.models.gemma3.gemma3_causal_lm_preprocessor",
37
+ "class_name": "Gemma3CausalLMPreprocessor",
38
+ "config": {
39
+ "name": "gemma3_causal_lm_preprocessor",
40
+ "trainable": true,
41
+ "dtype": {
42
+ "module": "keras",
43
+ "class_name": "DTypePolicy",
44
+ "config": {
45
+ "name": "float32"
46
+ },
47
+ "registered_name": null
48
+ },
49
+ "tokenizer": {
50
+ "module": "keras_hub.src.models.gemma3.gemma3_tokenizer",
51
+ "class_name": "Gemma3Tokenizer",
52
+ "config": {
53
+ "name": "gemma3_tokenizer",
54
+ "trainable": true,
55
+ "dtype": {
56
+ "module": "keras",
57
+ "class_name": "DTypePolicy",
58
+ "config": {
59
+ "name": "int32"
60
+ },
61
+ "registered_name": null
62
+ },
63
+ "config_file": "tokenizer.json",
64
+ "proto": null,
65
+ "sequence_length": null,
66
+ "add_bos": false,
67
+ "add_eos": false
68
+ },
69
+ "registered_name": "keras_hub>Gemma3Tokenizer"
70
+ },
71
+ "config_file": "preprocessor.json",
72
+ "sequence_length": 256,
73
+ "add_start_token": true,
74
+ "add_end_token": true,
75
+ "num_vision_tokens_per_image": 0,
76
+ "max_images_per_prompt": 0
77
+ },
78
+ "registered_name": "keras_hub>Gemma3CausalLMPreprocessor"
79
+ },
80
+ "name": "gemma3_causal_lm"
81
+ },
82
+ "registered_name": "keras_hub>Gemma3CausalLM"
83
+ }
tokenizer.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "module": "keras_hub.src.models.gemma3.gemma3_tokenizer",
3
+ "class_name": "Gemma3Tokenizer",
4
+ "config": {
5
+ "name": "gemma3_tokenizer",
6
+ "trainable": true,
7
+ "dtype": {
8
+ "module": "keras",
9
+ "class_name": "DTypePolicy",
10
+ "config": {
11
+ "name": "int32"
12
+ },
13
+ "registered_name": null
14
+ },
15
+ "config_file": "tokenizer.json",
16
+ "proto": null,
17
+ "sequence_length": null,
18
+ "add_bos": false,
19
+ "add_eos": false
20
+ },
21
+ "registered_name": "keras_hub>Gemma3Tokenizer"
22
+ }