Upload Qwen2VLForConditionalGeneration
Browse files- config.json +3 -29
- generation_config.json +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -4,6 +4,7 @@
|
|
| 4 |
],
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"bos_token_id": 151643,
|
|
|
|
| 7 |
"eos_token_id": 151645,
|
| 8 |
"hidden_act": "silu",
|
| 9 |
"hidden_size": 1536,
|
|
@@ -32,6 +33,7 @@
|
|
| 32 |
],
|
| 33 |
"attention_dropout": 0.0,
|
| 34 |
"bos_token_id": 151643,
|
|
|
|
| 35 |
"eos_token_id": 151645,
|
| 36 |
"hidden_act": "silu",
|
| 37 |
"hidden_size": 16,
|
|
@@ -39,32 +41,6 @@
|
|
| 39 |
"initializer_range": 0.02,
|
| 40 |
"intermediate_size": 8960,
|
| 41 |
"layer_types": [
|
| 42 |
-
"full_attention",
|
| 43 |
-
"full_attention",
|
| 44 |
-
"full_attention",
|
| 45 |
-
"full_attention",
|
| 46 |
-
"full_attention",
|
| 47 |
-
"full_attention",
|
| 48 |
-
"full_attention",
|
| 49 |
-
"full_attention",
|
| 50 |
-
"full_attention",
|
| 51 |
-
"full_attention",
|
| 52 |
-
"full_attention",
|
| 53 |
-
"full_attention",
|
| 54 |
-
"full_attention",
|
| 55 |
-
"full_attention",
|
| 56 |
-
"full_attention",
|
| 57 |
-
"full_attention",
|
| 58 |
-
"full_attention",
|
| 59 |
-
"full_attention",
|
| 60 |
-
"full_attention",
|
| 61 |
-
"full_attention",
|
| 62 |
-
"full_attention",
|
| 63 |
-
"full_attention",
|
| 64 |
-
"full_attention",
|
| 65 |
-
"full_attention",
|
| 66 |
-
"full_attention",
|
| 67 |
-
"full_attention",
|
| 68 |
"full_attention",
|
| 69 |
"full_attention"
|
| 70 |
],
|
|
@@ -85,7 +61,6 @@
|
|
| 85 |
"rope_theta": 1000000.0,
|
| 86 |
"sliding_window": null,
|
| 87 |
"tie_word_embeddings": true,
|
| 88 |
-
"torch_dtype": "bfloat16",
|
| 89 |
"use_cache": true,
|
| 90 |
"use_sliding_window": false,
|
| 91 |
"video_token_id": null,
|
|
@@ -94,8 +69,7 @@
|
|
| 94 |
"vision_token_id": 151654,
|
| 95 |
"vocab_size": 151936
|
| 96 |
},
|
| 97 |
-
"
|
| 98 |
-
"transformers_version": "4.56.0.dev0",
|
| 99 |
"use_cache": true,
|
| 100 |
"use_sliding_window": false,
|
| 101 |
"video_token_id": 151656,
|
|
|
|
| 4 |
],
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"bos_token_id": 151643,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
"eos_token_id": 151645,
|
| 9 |
"hidden_act": "silu",
|
| 10 |
"hidden_size": 1536,
|
|
|
|
| 33 |
],
|
| 34 |
"attention_dropout": 0.0,
|
| 35 |
"bos_token_id": 151643,
|
| 36 |
+
"dtype": "bfloat16",
|
| 37 |
"eos_token_id": 151645,
|
| 38 |
"hidden_act": "silu",
|
| 39 |
"hidden_size": 16,
|
|
|
|
| 41 |
"initializer_range": 0.02,
|
| 42 |
"intermediate_size": 8960,
|
| 43 |
"layer_types": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"full_attention",
|
| 45 |
"full_attention"
|
| 46 |
],
|
|
|
|
| 61 |
"rope_theta": 1000000.0,
|
| 62 |
"sliding_window": null,
|
| 63 |
"tie_word_embeddings": true,
|
|
|
|
| 64 |
"use_cache": true,
|
| 65 |
"use_sliding_window": false,
|
| 66 |
"video_token_id": null,
|
|
|
|
| 69 |
"vision_token_id": 151654,
|
| 70 |
"vocab_size": 151936
|
| 71 |
},
|
| 72 |
+
"transformers_version": "4.56.1",
|
|
|
|
| 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": "4.56.
|
| 6 |
}
|
|
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 151643,
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
+
"transformers_version": "4.56.1"
|
| 6 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 140919824
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2736fdf9b8bcb2acd00b2b333599f512dd0ec53f7dee0e8a37d93e6604fe308c
|
| 3 |
size 140919824
|