Upload training_logs/tsae_paper_k20__seed2.json with huggingface_hub
Browse files
training_logs/tsae_paper_k20__seed2.json
ADDED
|
@@ -0,0 +1,200 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"loss": [
|
| 3 |
+
112348.5078125,
|
| 4 |
+
24871.01953125,
|
| 5 |
+
11888.64453125,
|
| 6 |
+
8584.23828125,
|
| 7 |
+
7472.1328125,
|
| 8 |
+
7016.4921875,
|
| 9 |
+
6686.6708984375,
|
| 10 |
+
6276.4765625,
|
| 11 |
+
6171.8173828125,
|
| 12 |
+
5960.05224609375,
|
| 13 |
+
5924.404296875,
|
| 14 |
+
5785.9658203125,
|
| 15 |
+
5716.4443359375,
|
| 16 |
+
5679.7783203125,
|
| 17 |
+
5733.3642578125,
|
| 18 |
+
5629.67626953125,
|
| 19 |
+
5543.23095703125,
|
| 20 |
+
5472.462890625,
|
| 21 |
+
5481.11279296875,
|
| 22 |
+
5513.4521484375,
|
| 23 |
+
5440.625,
|
| 24 |
+
5375.6650390625,
|
| 25 |
+
5385.80224609375,
|
| 26 |
+
5313.6494140625,
|
| 27 |
+
5339.1806640625,
|
| 28 |
+
5279.513671875,
|
| 29 |
+
5322.462890625,
|
| 30 |
+
5362.87255859375,
|
| 31 |
+
5219.6181640625
|
| 32 |
+
],
|
| 33 |
+
"l0": [
|
| 34 |
+
20.0,
|
| 35 |
+
20.0,
|
| 36 |
+
20.0,
|
| 37 |
+
20.0,
|
| 38 |
+
20.0,
|
| 39 |
+
20.0,
|
| 40 |
+
20.0,
|
| 41 |
+
20.0,
|
| 42 |
+
20.0,
|
| 43 |
+
20.0,
|
| 44 |
+
20.0,
|
| 45 |
+
20.0,
|
| 46 |
+
20.0,
|
| 47 |
+
20.0,
|
| 48 |
+
20.0,
|
| 49 |
+
20.0,
|
| 50 |
+
20.0,
|
| 51 |
+
20.0,
|
| 52 |
+
20.0,
|
| 53 |
+
20.0,
|
| 54 |
+
20.0,
|
| 55 |
+
20.0,
|
| 56 |
+
20.0,
|
| 57 |
+
20.0,
|
| 58 |
+
20.0,
|
| 59 |
+
20.0,
|
| 60 |
+
20.0,
|
| 61 |
+
20.0,
|
| 62 |
+
20.0
|
| 63 |
+
],
|
| 64 |
+
"auxk": [
|
| 65 |
+
0.0,
|
| 66 |
+
0.0,
|
| 67 |
+
0.0,
|
| 68 |
+
0.0,
|
| 69 |
+
0.0,
|
| 70 |
+
0.0,
|
| 71 |
+
0.0,
|
| 72 |
+
0.0,
|
| 73 |
+
0.0,
|
| 74 |
+
0.0,
|
| 75 |
+
0.0,
|
| 76 |
+
0.0,
|
| 77 |
+
0.0,
|
| 78 |
+
0.0,
|
| 79 |
+
0.0,
|
| 80 |
+
0.0,
|
| 81 |
+
0.0,
|
| 82 |
+
0.0,
|
| 83 |
+
0.0,
|
| 84 |
+
1.000484824180603,
|
| 85 |
+
1.0002790689468384,
|
| 86 |
+
1.00020432472229,
|
| 87 |
+
0.999527096748352,
|
| 88 |
+
0.9983506202697754,
|
| 89 |
+
0.9961974620819092,
|
| 90 |
+
0.994956910610199,
|
| 91 |
+
0.9953610897064209,
|
| 92 |
+
0.9885474443435669,
|
| 93 |
+
0.9875771403312683
|
| 94 |
+
],
|
| 95 |
+
"dead": [
|
| 96 |
+
0,
|
| 97 |
+
0,
|
| 98 |
+
0,
|
| 99 |
+
0,
|
| 100 |
+
0,
|
| 101 |
+
0,
|
| 102 |
+
0,
|
| 103 |
+
0,
|
| 104 |
+
0,
|
| 105 |
+
0,
|
| 106 |
+
0,
|
| 107 |
+
0,
|
| 108 |
+
0,
|
| 109 |
+
0,
|
| 110 |
+
0,
|
| 111 |
+
0,
|
| 112 |
+
0,
|
| 113 |
+
0,
|
| 114 |
+
0,
|
| 115 |
+
8,
|
| 116 |
+
15,
|
| 117 |
+
40,
|
| 118 |
+
69,
|
| 119 |
+
127,
|
| 120 |
+
178,
|
| 121 |
+
236,
|
| 122 |
+
319,
|
| 123 |
+
428,
|
| 124 |
+
546
|
| 125 |
+
],
|
| 126 |
+
"steps_logged": [
|
| 127 |
+
0,
|
| 128 |
+
200,
|
| 129 |
+
400,
|
| 130 |
+
600,
|
| 131 |
+
800,
|
| 132 |
+
1000,
|
| 133 |
+
1200,
|
| 134 |
+
1400,
|
| 135 |
+
1600,
|
| 136 |
+
1800,
|
| 137 |
+
2000,
|
| 138 |
+
2200,
|
| 139 |
+
2400,
|
| 140 |
+
2600,
|
| 141 |
+
2800,
|
| 142 |
+
3000,
|
| 143 |
+
3200,
|
| 144 |
+
3400,
|
| 145 |
+
3600,
|
| 146 |
+
3800,
|
| 147 |
+
4000,
|
| 148 |
+
4200,
|
| 149 |
+
4400,
|
| 150 |
+
4600,
|
| 151 |
+
4800,
|
| 152 |
+
5000,
|
| 153 |
+
5200,
|
| 154 |
+
5400,
|
| 155 |
+
5600
|
| 156 |
+
],
|
| 157 |
+
"final_step": 5600,
|
| 158 |
+
"converged": true,
|
| 159 |
+
"plateau_last": 0.018703698722458955,
|
| 160 |
+
"elapsed_s": 570.1786015033722,
|
| 161 |
+
"group_sizes": [
|
| 162 |
+
3686,
|
| 163 |
+
14746
|
| 164 |
+
],
|
| 165 |
+
"group_weights": [
|
| 166 |
+
0.5,
|
| 167 |
+
0.5
|
| 168 |
+
],
|
| 169 |
+
"lr_used": 0.0001885618083164127,
|
| 170 |
+
"row": 3,
|
| 171 |
+
"arch_id": "tsae_paper_k20",
|
| 172 |
+
"arch": "tsae_paper_k20",
|
| 173 |
+
"group": 1,
|
| 174 |
+
"src_class": "TemporalMatryoshkaBatchTopKSAE",
|
| 175 |
+
"src_module": "src.architectures.tsae_paper",
|
| 176 |
+
"T": 1,
|
| 177 |
+
"T_max": null,
|
| 178 |
+
"t_sample": null,
|
| 179 |
+
"n_layers": null,
|
| 180 |
+
"k_win": 20,
|
| 181 |
+
"k_pos": 20,
|
| 182 |
+
"shifts": null,
|
| 183 |
+
"alpha": null,
|
| 184 |
+
"gamma": null,
|
| 185 |
+
"n_scales": null,
|
| 186 |
+
"seed": 2,
|
| 187 |
+
"d_in": 2304,
|
| 188 |
+
"d_sae": 18432,
|
| 189 |
+
"subject_model": "google/gemma-2-2b",
|
| 190 |
+
"anchor_layer": 12,
|
| 191 |
+
"mlc_layers": [
|
| 192 |
+
10,
|
| 193 |
+
11,
|
| 194 |
+
12,
|
| 195 |
+
13,
|
| 196 |
+
14
|
| 197 |
+
],
|
| 198 |
+
"phase": "phase7_unification",
|
| 199 |
+
"run_id": "tsae_paper_k20__seed2"
|
| 200 |
+
}
|