Fix mrope_section config: [2] -> [1, 1] for Liger kernel compatibility
Browse files- config.json +10 -10
- generation_config.json +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -7,20 +7,22 @@
|
|
| 7 |
"dtype": "bfloat16",
|
| 8 |
"eos_token_id": 151645,
|
| 9 |
"hidden_act": "silu",
|
| 10 |
-
"hidden_size":
|
| 11 |
"image_token_id": 151655,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"intermediate_size": 11008,
|
| 14 |
"max_position_embeddings": 128000,
|
| 15 |
"max_window_layers": 70,
|
| 16 |
"model_type": "qwen2_5_vl",
|
| 17 |
-
"num_attention_heads":
|
| 18 |
-
"num_hidden_layers":
|
| 19 |
"num_key_value_heads": 2,
|
| 20 |
"rms_norm_eps": 1e-06,
|
| 21 |
"rope_scaling": {
|
| 22 |
"mrope_section": [
|
| 23 |
-
|
|
|
|
|
|
|
| 24 |
],
|
| 25 |
"rope_type": "default",
|
| 26 |
"type": "default"
|
|
@@ -28,6 +30,7 @@
|
|
| 28 |
"rope_theta": 1000000.0,
|
| 29 |
"sliding_window": 32768,
|
| 30 |
"text_config": {
|
|
|
|
| 31 |
"architectures": [
|
| 32 |
"Qwen2_5_VLForConditionalGeneration"
|
| 33 |
],
|
|
@@ -37,7 +40,6 @@
|
|
| 37 |
"eos_token_id": 151645,
|
| 38 |
"hidden_act": "silu",
|
| 39 |
"hidden_size": 16,
|
| 40 |
-
"image_token_id": null,
|
| 41 |
"initializer_range": 0.02,
|
| 42 |
"intermediate_size": 11008,
|
| 43 |
"layer_types": [
|
|
@@ -53,7 +55,8 @@
|
|
| 53 |
"rms_norm_eps": 1e-06,
|
| 54 |
"rope_scaling": {
|
| 55 |
"mrope_section": [
|
| 56 |
-
|
|
|
|
| 57 |
],
|
| 58 |
"rope_type": "default",
|
| 59 |
"type": "default"
|
|
@@ -63,13 +66,10 @@
|
|
| 63 |
"tie_word_embeddings": true,
|
| 64 |
"use_cache": true,
|
| 65 |
"use_sliding_window": false,
|
| 66 |
-
"video_token_id": null,
|
| 67 |
-
"vision_end_token_id": 151653,
|
| 68 |
-
"vision_start_token_id": 151652,
|
| 69 |
"vision_token_id": 151654,
|
| 70 |
"vocab_size": 151936
|
| 71 |
},
|
| 72 |
-
"transformers_version": "4.
|
| 73 |
"use_cache": true,
|
| 74 |
"use_sliding_window": false,
|
| 75 |
"video_token_id": 151656,
|
|
|
|
| 7 |
"dtype": "bfloat16",
|
| 8 |
"eos_token_id": 151645,
|
| 9 |
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
"image_token_id": 151655,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"intermediate_size": 11008,
|
| 14 |
"max_position_embeddings": 128000,
|
| 15 |
"max_window_layers": 70,
|
| 16 |
"model_type": "qwen2_5_vl",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 36,
|
| 19 |
"num_key_value_heads": 2,
|
| 20 |
"rms_norm_eps": 1e-06,
|
| 21 |
"rope_scaling": {
|
| 22 |
"mrope_section": [
|
| 23 |
+
16,
|
| 24 |
+
24,
|
| 25 |
+
24
|
| 26 |
],
|
| 27 |
"rope_type": "default",
|
| 28 |
"type": "default"
|
|
|
|
| 30 |
"rope_theta": 1000000.0,
|
| 31 |
"sliding_window": 32768,
|
| 32 |
"text_config": {
|
| 33 |
+
"_name_or_path": "Qwen/Qwen2.5-VL-3B-Instruct",
|
| 34 |
"architectures": [
|
| 35 |
"Qwen2_5_VLForConditionalGeneration"
|
| 36 |
],
|
|
|
|
| 40 |
"eos_token_id": 151645,
|
| 41 |
"hidden_act": "silu",
|
| 42 |
"hidden_size": 16,
|
|
|
|
| 43 |
"initializer_range": 0.02,
|
| 44 |
"intermediate_size": 11008,
|
| 45 |
"layer_types": [
|
|
|
|
| 55 |
"rms_norm_eps": 1e-06,
|
| 56 |
"rope_scaling": {
|
| 57 |
"mrope_section": [
|
| 58 |
+
1,
|
| 59 |
+
1
|
| 60 |
],
|
| 61 |
"rope_type": "default",
|
| 62 |
"type": "default"
|
|
|
|
| 66 |
"tie_word_embeddings": true,
|
| 67 |
"use_cache": true,
|
| 68 |
"use_sliding_window": false,
|
|
|
|
|
|
|
|
|
|
| 69 |
"vision_token_id": 151654,
|
| 70 |
"vocab_size": 151936
|
| 71 |
},
|
| 72 |
+
"transformers_version": "4.57.1",
|
| 73 |
"use_cache": true,
|
| 74 |
"use_sliding_window": false,
|
| 75 |
"video_token_id": 151656,
|
generation_config.json
CHANGED
|
@@ -2,5 +2,5 @@
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 151643,
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
-
"transformers_version": "4.
|
| 6 |
}
|
|
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 151643,
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
+
"transformers_version": "4.57.1"
|
| 6 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 18086192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9ee6de012cbabb328744133306ae85256f9d1d994145c50a0a341918c7af35c
|
| 3 |
size 18086192
|