Hugging Face's logo

FlyinGodzilla
/
AMOR-Mamba2-1.5B

Text Generation
PyTorch
Safetensors
English
amor
custom-code
pretrained
research
mamba2
Model card Files
xet
 Community
AMOR-Mamba2-1.5B
File size: 4,636 Bytes
d95b776
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51b6330
 
d95b776
 
 
 
51b6330
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b66d9cc
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ee2b305
 
 
 
 
 
51b6330
d95b776
b66d9cc
 
 
 
 
 
 
 
ee2b305
b66d9cc
 
ee2b305
 
 
 
 
 
 
 
b66d9cc
d95b776
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
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
{
  "repo_id": "FlyinGodzilla/AMOR-Mamba2-1.5B",
  "source_filename": "amor_1p5b_step_0062558.pt",
  "source_sha256": "f3a96c227d5deddb2557de46328c937f75e06e85b058effe2b25294762d221dd",
  "step": 62558,
  "tokens_seen": 30748520448,
  "n_params": 1537425920,
  "tensor_dtypes": {
    "torch.float32": 338
  },
  "gate_buffers": {
    "0": {
      "running_threshold": 0.937967836856842,
      "running_std": 0.025595463812351227
    },
    "1": {
      "running_threshold": 0.9341728091239929,
      "running_std": 0.03567861020565033
    },
    "2": {
      "running_threshold": 0.93309086561203,
      "running_std": 0.028174862265586853
    }
  },
  "optimizer_included": false,
  "precision_changed": false,
  "validation": {
    "exported_tensors_equal": true,
    "strict_model_load": true,
    "generation": true,
    "tensor_names_and_shapes_match": true,
    "layout_validation_device": "meta",
    "layout_validation_torch": "2.6.0",
    "layout_validation_transformers": "4.51.3",
    "cuda": {
      "versions": {
        "torch": "2.6.0+cu124",
        "transformers": "5.0.0",
        "mamba-ssm": "2.3.0",
        "causal-conv1d": "1.6.0",
        "flash-linear-attention": "0.4.2",
        "triton": "3.2.0",
        "safetensors": "0.7.0"
      },
      "tokenizer_revision": "5590cc63198b0f111f4289daf006da0a12044173",
      "strict_model_load": true,
      "generation": true,
      "device": "NVIDIA H100 NVL",
      "torch": "2.6.0+cu124",
      "autocast": "bfloat16",
      "prefill_max_abs_diff": 0.0,
      "decode_max_abs_diff": 0.0625,
      "generated_token_ids": [
        264,
        3424,
        315,
        279,
        12632,
        1665,
        13,
        1102
      ],
      "published_cli_passed": true,
      "test_scope": "Short CUDA smoke test; not a benchmark rerun",
      "comparison_tolerance": {
        "rtol": 0.02,
        "atol": 0.1
      },
      "tested_decoder_sha256": "4b48db8e5fb67318244c533356903f40703156cdca24ccf4fd49086e72b16db7"
    },
    "router_release": {
      "exact_router_export": true,
      "base_model_association": true,
      "default_forward_max_abs_diff": 0.0,
      "modes": [
        {
          "router": false,
          "cases": [
            {
              "batch": 1,
              "length": 16,
              "prefill_max_abs_diff": 0.0625,
              "decode_max_abs_diff": 0.0625
            },
            {
              "batch": 1,
              "length": 128,
              "prefill_max_abs_diff": 0.0,
              "decode_max_abs_diff": 0.125
            },
            {
              "batch": 2,
              "length": 64,
              "prefill_max_abs_diff": 0.0
            }
          ],
          "canonical_router_exact": null
        },
        {
          "router": true,
          "cases": [
            {
              "batch": 1,
              "length": 16,
              "prefill_max_abs_diff": 0.0625,
              "decode_max_abs_diff": 0.0625
            },
            {
              "batch": 1,
              "length": 128,
              "prefill_max_abs_diff": 0.0,
              "decode_max_abs_diff": 0.0625
            },
            {
              "batch": 2,
              "length": 64,
              "prefill_max_abs_diff": 0.0
            }
          ],
          "canonical_router_exact": true
        }
      ],
      "frozen_thresholds_preserved": true,
      "training_router_rejected": true,
      "gpu": "NVIDIA H100 NVL",
      "torch": "2.6.0+cu124",
      "generation_cli_modes_passed": [
        "entropy",
        "router"
      ]
    },
    "current_release": {
      "strict_model_load": true,
      "entropy_and_router_generation_passed": true,
      "unit_tests_passed": 17,
      "device": "NVIDIA H100 NVL"
    }
  },
  "implementation": "AMOR shared release: native entropy gate and optional fitted inference router",
  "router": {
    "variant": "distill",
    "hidden_dim": 512,
    "activation": "silu",
    "weights": "router.safetensors",
    "config": "router_config.json",
    "sha256": "e600c3dc0f10fb1460e7c4ccbf2a5a5d97b18b43c509dfdff6b932e2f36601d4",
    "base_weight_sha256": "e395551632002619ab2f65aa42e7ab16f4ce0b56e4e75080be726b2331c8b19d",
    "precision": "float32",
    "inference_only": true
  },
  "weight_layout": {
    "format": "safetensors-sharded-v1",
    "index": "model.safetensors.index.json",
    "shards": 8,
    "weight_identity_sha256": "e395551632002619ab2f65aa42e7ab16f4ce0b56e4e75080be726b2331c8b19d",
    "original_single_file_sha256": "878f57056e17113f01798bd3e542d6e59a9e7a6067f994aeab726808bca6bfe4",
    "all_tensor_bytes_identical": true
  }
}