TomasFAV commited on
Commit
c8fc681
·
1 Parent(s): eace4c7

Training in progress, epoch 1

Browse files
config.json CHANGED
@@ -19,16 +19,18 @@
19
  "d_kv": 64,
20
  "decoder_start_token_id": 0,
21
  "dense_act_fn": "gelu_new",
22
- "dropout_rate": 0.2,
23
  "dtype": "float32",
24
  "encoder_hidden_size": 768,
25
  "eos_token_id": 1,
 
26
  "finetuning_task": null,
27
  "hidden_size": 768,
28
  "initializer_factor": 1.0,
29
  "initializer_range": 0.02,
30
  "is_decoder": true,
31
  "is_encoder_decoder": true,
 
32
  "layer_norm_epsilon": 1e-06,
33
  "model_type": "pix2struct_text_model",
34
  "num_heads": 12,
@@ -54,24 +56,26 @@
54
  "use_cache": false,
55
  "vision_config": {
56
  "add_cross_attention": false,
57
- "attention_dropout": 0.2,
58
  "bos_token_id": null,
59
  "cross_attention_hidden_size": null,
60
  "d_ff": 2048,
61
  "d_kv": 64,
62
  "decoder_start_token_id": null,
63
  "dense_act_fn": "gelu_new",
64
- "dropout_rate": 0.2,
65
  "dtype": "float32",
66
  "eos_token_id": null,
67
  "finetuning_task": null,
68
- "hidden_dropout_prob": 0.2,
69
  "hidden_size": 768,
 
70
  "initializer_factor": 1.0,
71
  "initializer_range": 0.02,
72
  "is_decoder": false,
73
  "layer_norm_bias": false,
74
  "layer_norm_eps": 1e-06,
 
75
  "model_type": "pix2struct_vision_model",
76
  "num_attention_heads": 12,
77
  "num_channels": 3,
@@ -82,6 +86,7 @@
82
  "prefix": null,
83
  "projection_dim": 768,
84
  "pruned_heads": {},
 
85
  "relative_attention_max_distance": 128,
86
  "relative_attention_num_buckets": 32,
87
  "sep_token_id": null,
 
19
  "d_kv": 64,
20
  "decoder_start_token_id": 0,
21
  "dense_act_fn": "gelu_new",
22
+ "dropout_rate": 0.05,
23
  "dtype": "float32",
24
  "encoder_hidden_size": 768,
25
  "eos_token_id": 1,
26
+ "feed_forward_proj": "gated-gelu",
27
  "finetuning_task": null,
28
  "hidden_size": 768,
29
  "initializer_factor": 1.0,
30
  "initializer_range": 0.02,
31
  "is_decoder": true,
32
  "is_encoder_decoder": true,
33
+ "is_gated_act": true,
34
  "layer_norm_epsilon": 1e-06,
35
  "model_type": "pix2struct_text_model",
36
  "num_heads": 12,
 
56
  "use_cache": false,
57
  "vision_config": {
58
  "add_cross_attention": false,
59
+ "attention_dropout": 0.05,
60
  "bos_token_id": null,
61
  "cross_attention_hidden_size": null,
62
  "d_ff": 2048,
63
  "d_kv": 64,
64
  "decoder_start_token_id": null,
65
  "dense_act_fn": "gelu_new",
66
+ "dropout_rate": 0.06,
67
  "dtype": "float32",
68
  "eos_token_id": null,
69
  "finetuning_task": null,
70
+ "hidden_dropout_prob": 0.05,
71
  "hidden_size": 768,
72
+ "image_size": 384,
73
  "initializer_factor": 1.0,
74
  "initializer_range": 0.02,
75
  "is_decoder": false,
76
  "layer_norm_bias": false,
77
  "layer_norm_eps": 1e-06,
78
+ "mlp_bias": false,
79
  "model_type": "pix2struct_vision_model",
80
  "num_attention_heads": 12,
81
  "num_channels": 3,
 
86
  "prefix": null,
87
  "projection_dim": 768,
88
  "pruned_heads": {},
89
+ "qkv_bias": false,
90
  "relative_attention_max_distance": 128,
91
  "relative_attention_num_buckets": 32,
92
  "sep_token_id": null,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aec55bcb6eb7b1266535a4027ae78f875d7dcbbb1c783d75e177848940e3dc5a
3
  size 1130333048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42fba9164811de2a977ad5991bf1852c6b3641f7d9b1c293ebb2d2a03ea47b5a
3
  size 1130333048
runs/Mar27_17-54-26_b4d31b44522d/events.out.tfevents.1774634066.b4d31b44522d.1455.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9482e1fec613ee028a915d8bc930baacbf4610711ed8bead5268a01d594d3f61
3
+ size 10088
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:933aa010c21f5a53cd50f151d4595d8afe82f11a2c8ca162b294a81b7a17f92f
3
  size 5393
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56eae8dcd443552f14850f7cd98c1a264a724cceaaa5c9febfd097023d41119f
3
  size 5393