Hugging Face's logo

FredrikKarlssonSpeech
/
megatron-bert-base-swedish-cased-600k-onnx

Feature Extraction
ONNX
Swedish
onnxruntime
swedish
bert
fill-mask
megatron-bert
Model card Files
xet
 Community
megatron-bert-base-swedish-cased-600k-onnx
File size: 1,580 Bytes
0d78da3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
{
  "model_id": "KBLab/megatron-bert-base-swedish-cased-600k",
  "fill-mask": {
    "shape": [
      1,
      10,
      64128
    ],
    "max_abs_diff": 6.884336471557617e-05,
    "mean_abs_diff": 8.68798088049516e-06,
    "input_text": "Det svenska ordet for kung ar [MASK].",
    "mask_index": 7,
    "pytorch": {
      "logits_shape": [
        1,
        10,
        64128
      ],
      "top5_token_ids": [
        37090,
        57980,
        35856,
        50325,
        55930
      ],
      "top5_tokens": [
        "kung",
        "hertig",
        "fe",
        "bonde",
        "konung"
      ]
    },
    "onnx_fp32": {
      "logits_shape": [
        1,
        10,
        64128
      ],
      "top5_token_ids": [
        37090,
        57980,
        35856,
        50325,
        55930
      ],
      "top5_tokens": [
        "kung",
        "hertig",
        "fe",
        "bonde",
        "konung"
      ]
    }
  },
  "feature-extraction": {
    "input_text": "Svensk BERT ska kunna skapa embeddings for den har texten.",
    "last_hidden_state": {
      "pytorch_shape": [
        1,
        18,
        768
      ],
      "onnx_fp32_shape": [
        1,
        18,
        768
      ],
      "max_abs_diff": 5.930662155151367e-06,
      "mean_abs_diff": 1.0840861932592816e-06
    },
    "cls_embedding": {
      "pytorch_shape": [
        1,
        768
      ],
      "onnx_fp32_shape": [
        1,
        768
      ],
      "max_abs_diff": 2.950429916381836e-06,
      "mean_abs_diff": 5.458733198793198e-07,
      "cosine_similarity": 1.0
    }
  }
}