Hugging Face's logo

zhiyang1
/
YG_vlm_dc-vae_layerwise_ar_steps-1_img_diffuser-768-12

Safetensors
Model card Files
xet
Community
zhiyang1 commited on
Commit
aa9f12d
·
verified ·
1 Parent(s): 492ffef

Upload folder using huggingface_hub

Browse files
Files changed (15) hide show
  1. checkpoint-20000/config.json +71 -0
  2. checkpoint-20000/generation_config.json +6 -0
  3. checkpoint-20000/model.safetensors +3 -0
  4. checkpoint-20000/optimizer.pt +3 -0
  5. checkpoint-20000/rng_state_0.pth +3 -0
  6. checkpoint-20000/rng_state_1.pth +3 -0
  7. checkpoint-20000/rng_state_2.pth +3 -0
  8. checkpoint-20000/rng_state_3.pth +3 -0
  9. checkpoint-20000/rng_state_4.pth +3 -0
  10. checkpoint-20000/rng_state_5.pth +3 -0
  11. checkpoint-20000/rng_state_6.pth +3 -0
  12. checkpoint-20000/rng_state_7.pth +3 -0
  13. checkpoint-20000/scheduler.pt +3 -0
  14. checkpoint-20000/trainer_state.json +0 -0
  15. checkpoint-20000/training_args.bin +3 -0
checkpoint-20000/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "ar_steps": 1,
3
+ "architectures": [
4
+ "DiffVLMDiffusion"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "image_token_id": 151655,
12
+ "img_cross_attention_dim": 2048,
13
+ "img_diffuser_depth": 12,
14
+ "img_ffn_dim_multiplier": null,
15
+ "img_hidden_size": 768,
16
+ "img_multiple_of": 256,
17
+ "img_norm_eps": 1e-05,
18
+ "img_num_attention_heads": 12,
19
+ "img_num_kv_heads": 12,
20
+ "img_qk_norm": true,
21
+ "in_channels": 32,
22
+ "initializer_range": 0.02,
23
+ "inject_img_diffuser": true,
24
+ "input_size": 32,
25
+ "intermediate_size": 8960,
26
+ "lora_alpha": 128,
27
+ "lora_bias": "none",
28
+ "lora_dropout": 0.05,
29
+ "lora_enable": false,
30
+ "lora_r": 64,
31
+ "max_position_embeddings": 32768,
32
+ "max_window_layers": 28,
33
+ "model_type": "qwen2_vl",
34
+ "non_linearity": 1,
35
+ "norm_elementwise_affine": true,
36
+ "num_attention_heads": 12,
37
+ "num_hidden_layers": 28,
38
+ "num_key_value_heads": 2,
39
+ "patch_size": 2,
40
+ "rms_norm_eps": 1e-06,
41
+ "rope_scaling": {
42
+ "mrope_section": [
43
+ 16,
44
+ 24,
45
+ 24
46
+ ],
47
+ "rope_type": "default",
48
+ "type": "default"
49
+ },
50
+ "rope_theta": 1000000.0,
51
+ "sample_size": 128,
52
+ "sampling_steps": 28,
53
+ "sliding_window": null,
54
+ "tie_word_embeddings": true,
55
+ "torch_dtype": "bfloat16",
56
+ "transformers_version": "4.47.0",
57
+ "use_cache": true,
58
+ "use_sliding_window": false,
59
+ "vae_path": "mit-han-lab/dc-ae-f32c32-sana-1.1-diffusers",
60
+ "video_token_id": 151656,
61
+ "vision_config": {
62
+ "hidden_size": 1536,
63
+ "in_chans": 3,
64
+ "model_type": "qwen2_vl",
65
+ "spatial_patch_size": 14
66
+ },
67
+ "vision_end_token_id": 151653,
68
+ "vision_start_token_id": 151652,
69
+ "vision_token_id": 151654,
70
+ "vocab_size": 151936
71
+ }
checkpoint-20000/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151645,
5
+ "transformers_version": "4.47.0"
6
+ }
checkpoint-20000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f641bf8c42053b6405ff0cd6b4447a39d70856ecd85674e4ab92e3c2cef8e5a
3
+ size 4699255646
checkpoint-20000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:550891d88e3d6a3357ac5a48cf5baad68155d1741b3b76b92414a270d507b8f5
3
+ size 6750054312
checkpoint-20000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc090dbed776d53e97d7006269e5a2acac8a1d1cb2c81cd9df57b8cd2e90e8e2
3
+ size 15920
checkpoint-20000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ac9868f9e68bf06ea39568f4b551002a07dfe670732054e223b518cbaa2d5f8
3
+ size 15920
checkpoint-20000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a7d9177b00f8ab5dca55af697b46e57f913873c90ba190bfbc5104d67818e81
3
+ size 15920
checkpoint-20000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87b53b3757f22c39224860520c0960dcee407658ecbbe51f93627023f73005a5
3
+ size 15920
checkpoint-20000/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9db8a1523672356fd7d8b8b86fc66e06059c495a9662cb6f872410cff4e7c207
3
+ size 15920
checkpoint-20000/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9c7f92e93afc50136e5e78d6c84ee3c1f1d46e3340a6cbe39a457578ea7cc0a
3
+ size 15920
checkpoint-20000/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:286a6c74757d8cc02be7911dfd181e4b5d75cc24327ef8724a7ef07199a51676
3
+ size 15920
checkpoint-20000/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65aae95ed4c518b7b3b2233ec726733b078195253011c9d6f7007eba80b133a2
3
+ size 15920
checkpoint-20000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f03f24b02bdd285f31b8705583130ff132519634aa9400055ed20b5470b95107
3
+ size 1064
checkpoint-20000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-20000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf05db7c34795e323da49046c1c54494c736d1b6425d350630ab2fb6ee58b235
3
+ size 5880