Fix: upload medium size model instead of small
#2
by
Lauler
- opened
- config.json +20 -36
- generation_config.json +1 -1
- model.safetensors +2 -2
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "/leonardo_work/EUHPC_A01_006/experiments_whisper/stage1_results/
|
3 |
"activation_dropout": 0.1,
|
4 |
"activation_function": "gelu",
|
5 |
"apply_spec_augment": false,
|
@@ -10,17 +10,17 @@
|
|
10 |
"begin_suppress_tokens": null,
|
11 |
"bos_token_id": 50257,
|
12 |
"classifier_proj_size": 256,
|
13 |
-
"d_model":
|
14 |
-
"decoder_attention_heads":
|
15 |
-
"decoder_ffn_dim":
|
16 |
"decoder_layerdrop": 0.0,
|
17 |
-
"decoder_layers":
|
18 |
"decoder_start_token_id": 50258,
|
19 |
"dropout": 0.0,
|
20 |
-
"encoder_attention_heads":
|
21 |
-
"encoder_ffn_dim":
|
22 |
"encoder_layerdrop": 0.0,
|
23 |
-
"encoder_layers":
|
24 |
"eos_token_id": 50257,
|
25 |
"forced_decoder_ids": null,
|
26 |
"init_std": 0.02,
|
@@ -36,7 +36,7 @@
|
|
36 |
"max_target_positions": 448,
|
37 |
"median_filter_width": 7,
|
38 |
"model_type": "whisper",
|
39 |
-
"num_hidden_layers":
|
40 |
"num_mel_bins": 80,
|
41 |
"pad_token_id": 50257,
|
42 |
"scale_embedding": false,
|
@@ -47,44 +47,28 @@
|
|
47 |
"vocab_size": 51865,
|
48 |
"alignment_heads": [
|
49 |
[
|
50 |
-
|
51 |
-
|
52 |
],
|
53 |
[
|
54 |
-
|
55 |
-
9
|
56 |
-
],
|
57 |
-
[
|
58 |
-
8,
|
59 |
-
0
|
60 |
-
],
|
61 |
-
[
|
62 |
-
8,
|
63 |
4
|
64 |
],
|
65 |
[
|
66 |
-
|
67 |
-
|
68 |
],
|
69 |
[
|
70 |
-
|
71 |
-
|
72 |
],
|
73 |
[
|
74 |
-
|
75 |
0
|
76 |
],
|
77 |
[
|
78 |
-
|
79 |
-
|
80 |
-
],
|
81 |
-
[
|
82 |
-
9,
|
83 |
-
9
|
84 |
-
],
|
85 |
-
[
|
86 |
-
10,
|
87 |
-
5
|
88 |
]
|
89 |
],
|
90 |
"lang_ids": [
|
@@ -282,4 +266,4 @@
|
|
282 |
220,
|
283 |
50257
|
284 |
]
|
285 |
-
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "/leonardo_work/EUHPC_A01_006/experiments_whisper/stage1_results/medium_250210_124500/",
|
3 |
"activation_dropout": 0.1,
|
4 |
"activation_function": "gelu",
|
5 |
"apply_spec_augment": false,
|
|
|
10 |
"begin_suppress_tokens": null,
|
11 |
"bos_token_id": 50257,
|
12 |
"classifier_proj_size": 256,
|
13 |
+
"d_model": 1024,
|
14 |
+
"decoder_attention_heads": 16,
|
15 |
+
"decoder_ffn_dim": 4096,
|
16 |
"decoder_layerdrop": 0.0,
|
17 |
+
"decoder_layers": 24,
|
18 |
"decoder_start_token_id": 50258,
|
19 |
"dropout": 0.0,
|
20 |
+
"encoder_attention_heads": 16,
|
21 |
+
"encoder_ffn_dim": 4096,
|
22 |
"encoder_layerdrop": 0.0,
|
23 |
+
"encoder_layers": 24,
|
24 |
"eos_token_id": 50257,
|
25 |
"forced_decoder_ids": null,
|
26 |
"init_std": 0.02,
|
|
|
36 |
"max_target_positions": 448,
|
37 |
"median_filter_width": 7,
|
38 |
"model_type": "whisper",
|
39 |
+
"num_hidden_layers": 24,
|
40 |
"num_mel_bins": 80,
|
41 |
"pad_token_id": 50257,
|
42 |
"scale_embedding": false,
|
|
|
47 |
"vocab_size": 51865,
|
48 |
"alignment_heads": [
|
49 |
[
|
50 |
+
13,
|
51 |
+
15
|
52 |
],
|
53 |
[
|
54 |
+
15,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
4
|
56 |
],
|
57 |
[
|
58 |
+
15,
|
59 |
+
15
|
60 |
],
|
61 |
[
|
62 |
+
16,
|
63 |
+
1
|
64 |
],
|
65 |
[
|
66 |
+
20,
|
67 |
0
|
68 |
],
|
69 |
[
|
70 |
+
23,
|
71 |
+
4
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
72 |
]
|
73 |
],
|
74 |
"lang_ids": [
|
|
|
266 |
220,
|
267 |
50257
|
268 |
]
|
269 |
+
}
|
generation_config.json
CHANGED
@@ -237,4 +237,4 @@
|
|
237 |
"translate": 50358
|
238 |
},
|
239 |
"transformers_version": "4.45.2"
|
240 |
-
}
|
|
|
237 |
"translate": 50358
|
238 |
},
|
239 |
"transformers_version": "4.45.2"
|
240 |
+
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5d9eaecc00a1ca51b8f5e5f3da6270b4d234b5d3c94e1015316059a644d0320
|
3 |
+
size 1634047376
|