MHC-A1 durable progress index
Browse files
experiments/MHC_REAL_A1/progress.json
CHANGED
|
@@ -7,8 +7,8 @@
|
|
| 7 |
"arm": "stateless",
|
| 8 |
"validation": {
|
| 9 |
"W3_MSE": 0.11103641552229722,
|
| 10 |
-
"W4_MSE": 0.
|
| 11 |
-
"ctrl_max_rms": 0.
|
| 12 |
},
|
| 13 |
"checkpoint_step": 64
|
| 14 |
},
|
|
@@ -18,7 +18,7 @@
|
|
| 18 |
"validation": {
|
| 19 |
"W3_MSE": 0.11103636150558789,
|
| 20 |
"W4_MSE": 0.10791248766084512,
|
| 21 |
-
"ctrl_max_rms": 0.
|
| 22 |
},
|
| 23 |
"checkpoint_step": 64
|
| 24 |
}
|
|
@@ -30,7 +30,7 @@
|
|
| 30 |
"step": 0,
|
| 31 |
"loss_train": 0.09119217842817307,
|
| 32 |
"grad_norm": 0.013730722479522228,
|
| 33 |
-
"head_grad_norm": 1.
|
| 34 |
"q0_trials": [
|
| 35 |
{
|
| 36 |
"radius": 0.0003,
|
|
@@ -64,7 +64,7 @@
|
|
| 64 |
"loss_train": 0.08606645464897156,
|
| 65 |
"W3_MSE": 0.1147225337723891,
|
| 66 |
"W4_MSE": 0.11396808549761772,
|
| 67 |
-
"ctrl_max_rms": 0.
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"seed": 17,
|
|
@@ -73,7 +73,7 @@
|
|
| 73 |
"loss_train": 0.1242414116859436,
|
| 74 |
"W3_MSE": 0.11465953538815181,
|
| 75 |
"W4_MSE": 0.11387925470868747,
|
| 76 |
-
"ctrl_max_rms": 0.
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"seed": 17,
|
|
@@ -81,8 +81,8 @@
|
|
| 81 |
"step": 24,
|
| 82 |
"loss_train": 0.09662917256355286,
|
| 83 |
"W3_MSE": 0.11453648408253987,
|
| 84 |
-
"W4_MSE": 0.
|
| 85 |
-
"ctrl_max_rms": 0.
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"seed": 17,
|
|
@@ -91,7 +91,7 @@
|
|
| 91 |
"loss_train": 0.1582513302564621,
|
| 92 |
"W3_MSE": 0.1143090259283781,
|
| 93 |
"W4_MSE": 0.11331315711140633,
|
| 94 |
-
"ctrl_max_rms": 0.
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"seed": 17,
|
|
@@ -108,15 +108,15 @@
|
|
| 108 |
"step": 48,
|
| 109 |
"loss_train": 0.10963404178619385,
|
| 110 |
"W3_MSE": 0.1130196029941241,
|
| 111 |
-
"W4_MSE": 0.
|
| 112 |
-
"ctrl_max_rms": 0.
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"seed": 17,
|
| 116 |
"arm": "stateless",
|
| 117 |
"step": 56,
|
| 118 |
"loss_train": 0.10562591254711151,
|
| 119 |
-
"W3_MSE": 0.
|
| 120 |
"W4_MSE": 0.10920579296847184,
|
| 121 |
"ctrl_max_rms": 0.004299115234365066
|
| 122 |
},
|
|
@@ -126,8 +126,8 @@
|
|
| 126 |
"step": 64,
|
| 127 |
"loss_train": 0.1330445259809494,
|
| 128 |
"W3_MSE": 0.11103641552229722,
|
| 129 |
-
"W4_MSE": 0.
|
| 130 |
-
"ctrl_max_rms": 0.
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"seed": 17,
|
|
@@ -167,9 +167,9 @@
|
|
| 167 |
"arm": "mhc",
|
| 168 |
"step": 8,
|
| 169 |
"loss_train": 0.08606645464897156,
|
| 170 |
-
"W3_MSE": 0.
|
| 171 |
"W4_MSE": 0.113968084876736,
|
| 172 |
-
"ctrl_max_rms": 0.
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"seed": 17,
|
|
@@ -178,7 +178,7 @@
|
|
| 178 |
"loss_train": 0.1242414116859436,
|
| 179 |
"W3_MSE": 0.11465953538815181,
|
| 180 |
"W4_MSE": 0.11387925657133262,
|
| 181 |
-
"ctrl_max_rms": 0.
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"seed": 17,
|
|
@@ -187,7 +187,7 @@
|
|
| 187 |
"loss_train": 0.09662917256355286,
|
| 188 |
"W3_MSE": 0.11453648408253987,
|
| 189 |
"W4_MSE": 0.11368757796784242,
|
| 190 |
-
"ctrl_max_rms": 0.
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"seed": 17,
|
|
@@ -196,7 +196,7 @@
|
|
| 196 |
"loss_train": 0.1582513302564621,
|
| 197 |
"W3_MSE": 0.11430902406573296,
|
| 198 |
"W4_MSE": 0.11331315897405148,
|
| 199 |
-
"ctrl_max_rms": 0.
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"seed": 17,
|
|
@@ -205,25 +205,25 @@
|
|
| 205 |
"loss_train": 0.10655954480171204,
|
| 206 |
"W3_MSE": 0.11380419631799062,
|
| 207 |
"W4_MSE": 0.11249386705458164,
|
| 208 |
-
"ctrl_max_rms": 0.
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"seed": 17,
|
| 212 |
"arm": "mhc",
|
| 213 |
"step": 48,
|
| 214 |
"loss_train": 0.109634168446064,
|
| 215 |
-
"W3_MSE": 0.
|
| 216 |
"W4_MSE": 0.11111390093962352,
|
| 217 |
-
"ctrl_max_rms": 0.
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"seed": 17,
|
| 221 |
"arm": "mhc",
|
| 222 |
"step": 56,
|
| 223 |
"loss_train": 0.10562613606452942,
|
| 224 |
-
"W3_MSE": 0.
|
| 225 |
-
"W4_MSE": 0.
|
| 226 |
-
"ctrl_max_rms": 0.
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"seed": 17,
|
|
@@ -232,7 +232,7 @@
|
|
| 232 |
"loss_train": 0.13304445147514343,
|
| 233 |
"W3_MSE": 0.11103636150558789,
|
| 234 |
"W4_MSE": 0.10791248766084512,
|
| 235 |
-
"ctrl_max_rms": 0.
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"seed": 17,
|
|
@@ -275,42 +275,42 @@
|
|
| 275 |
"bytes": 93186
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"sha256": "
|
| 279 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0008.pt",
|
| 280 |
"bytes": 268976
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"sha256": "
|
| 284 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0016.pt",
|
| 285 |
"bytes": 268976
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"sha256": "
|
| 289 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0024.pt",
|
| 290 |
"bytes": 268976
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"sha256": "
|
| 294 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0032.pt",
|
| 295 |
"bytes": 268976
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"sha256": "
|
| 299 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0040.pt",
|
| 300 |
"bytes": 268976
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"sha256": "
|
| 304 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0048.pt",
|
| 305 |
"bytes": 268976
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"sha256": "
|
| 309 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0056.pt",
|
| 310 |
"bytes": 268976
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"sha256": "
|
| 314 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0064.pt",
|
| 315 |
"bytes": 268976
|
| 316 |
},
|
|
|
|
| 7 |
"arm": "stateless",
|
| 8 |
"validation": {
|
| 9 |
"W3_MSE": 0.11103641552229722,
|
| 10 |
+
"W4_MSE": 0.10791260438660781,
|
| 11 |
+
"ctrl_max_rms": 0.00528129100954781
|
| 12 |
},
|
| 13 |
"checkpoint_step": 64
|
| 14 |
},
|
|
|
|
| 18 |
"validation": {
|
| 19 |
"W3_MSE": 0.11103636150558789,
|
| 20 |
"W4_MSE": 0.10791248766084512,
|
| 21 |
+
"ctrl_max_rms": 0.005226058148158093
|
| 22 |
},
|
| 23 |
"checkpoint_step": 64
|
| 24 |
}
|
|
|
|
| 30 |
"step": 0,
|
| 31 |
"loss_train": 0.09119217842817307,
|
| 32 |
"grad_norm": 0.013730722479522228,
|
| 33 |
+
"head_grad_norm": 1.4221210449250066e-06,
|
| 34 |
"q0_trials": [
|
| 35 |
{
|
| 36 |
"radius": 0.0003,
|
|
|
|
| 64 |
"loss_train": 0.08606645464897156,
|
| 65 |
"W3_MSE": 0.1147225337723891,
|
| 66 |
"W4_MSE": 0.11396808549761772,
|
| 67 |
+
"ctrl_max_rms": 0.00018752604228211567
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"seed": 17,
|
|
|
|
| 73 |
"loss_train": 0.1242414116859436,
|
| 74 |
"W3_MSE": 0.11465953538815181,
|
| 75 |
"W4_MSE": 0.11387925470868747,
|
| 76 |
+
"ctrl_max_rms": 0.00030438662846184644
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"seed": 17,
|
|
|
|
| 81 |
"step": 24,
|
| 82 |
"loss_train": 0.09662917256355286,
|
| 83 |
"W3_MSE": 0.11453648408253987,
|
| 84 |
+
"W4_MSE": 0.11368757920960586,
|
| 85 |
+
"ctrl_max_rms": 0.0003923155600205064
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"seed": 17,
|
|
|
|
| 91 |
"loss_train": 0.1582513302564621,
|
| 92 |
"W3_MSE": 0.1143090259283781,
|
| 93 |
"W4_MSE": 0.11331315711140633,
|
| 94 |
+
"ctrl_max_rms": 0.0005296767340041697
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"seed": 17,
|
|
|
|
| 108 |
"step": 48,
|
| 109 |
"loss_train": 0.10963404178619385,
|
| 110 |
"W3_MSE": 0.1130196029941241,
|
| 111 |
+
"W4_MSE": 0.11111373826861382,
|
| 112 |
+
"ctrl_max_rms": 0.00205805353471078
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"seed": 17,
|
| 116 |
"arm": "stateless",
|
| 117 |
"step": 56,
|
| 118 |
"loss_train": 0.10562591254711151,
|
| 119 |
+
"W3_MSE": 0.11191886352996032,
|
| 120 |
"W4_MSE": 0.10920579296847184,
|
| 121 |
"ctrl_max_rms": 0.004299115234365066
|
| 122 |
},
|
|
|
|
| 126 |
"step": 64,
|
| 127 |
"loss_train": 0.1330445259809494,
|
| 128 |
"W3_MSE": 0.11103641552229722,
|
| 129 |
+
"W4_MSE": 0.10791260438660781,
|
| 130 |
+
"ctrl_max_rms": 0.00528129100954781
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"seed": 17,
|
|
|
|
| 167 |
"arm": "mhc",
|
| 168 |
"step": 8,
|
| 169 |
"loss_train": 0.08606645464897156,
|
| 170 |
+
"W3_MSE": 0.11472253315150738,
|
| 171 |
"W4_MSE": 0.113968084876736,
|
| 172 |
+
"ctrl_max_rms": 0.00018755901813468276
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"seed": 17,
|
|
|
|
| 178 |
"loss_train": 0.1242414116859436,
|
| 179 |
"W3_MSE": 0.11465953538815181,
|
| 180 |
"W4_MSE": 0.11387925657133262,
|
| 181 |
+
"ctrl_max_rms": 0.0003044217810383998
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"seed": 17,
|
|
|
|
| 187 |
"loss_train": 0.09662917256355286,
|
| 188 |
"W3_MSE": 0.11453648408253987,
|
| 189 |
"W4_MSE": 0.11368757796784242,
|
| 190 |
+
"ctrl_max_rms": 0.00039232581184478477
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"seed": 17,
|
|
|
|
| 196 |
"loss_train": 0.1582513302564621,
|
| 197 |
"W3_MSE": 0.11430902406573296,
|
| 198 |
"W4_MSE": 0.11331315897405148,
|
| 199 |
+
"ctrl_max_rms": 0.0005288315539170677
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"seed": 17,
|
|
|
|
| 205 |
"loss_train": 0.10655954480171204,
|
| 206 |
"W3_MSE": 0.11380419631799062,
|
| 207 |
"W4_MSE": 0.11249386705458164,
|
| 208 |
+
"ctrl_max_rms": 0.0009341221496773263
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"seed": 17,
|
| 212 |
"arm": "mhc",
|
| 213 |
"step": 48,
|
| 214 |
"loss_train": 0.109634168446064,
|
| 215 |
+
"W3_MSE": 0.11301966508229573,
|
| 216 |
"W4_MSE": 0.11111390093962352,
|
| 217 |
+
"ctrl_max_rms": 0.0020173621790794036
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"seed": 17,
|
| 221 |
"arm": "mhc",
|
| 222 |
"step": 56,
|
| 223 |
"loss_train": 0.10562613606452942,
|
| 224 |
+
"W3_MSE": 0.1119188933322827,
|
| 225 |
+
"W4_MSE": 0.10920589851836364,
|
| 226 |
+
"ctrl_max_rms": 0.004241040248113374
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"seed": 17,
|
|
|
|
| 232 |
"loss_train": 0.13304445147514343,
|
| 233 |
"W3_MSE": 0.11103636150558789,
|
| 234 |
"W4_MSE": 0.10791248766084512,
|
| 235 |
+
"ctrl_max_rms": 0.005226058148158093
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"seed": 17,
|
|
|
|
| 275 |
"bytes": 93186
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"sha256": "8564fa6b02b9004df33b139b15888e345f27bc9392db5b1bdd36918417abb5d3",
|
| 279 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0008.pt",
|
| 280 |
"bytes": 268976
|
| 281 |
},
|
| 282 |
{
|
| 283 |
+
"sha256": "42792020756dbc501ab231c407134a90b13db10de7d9dca0192c4da6eabafbae",
|
| 284 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0016.pt",
|
| 285 |
"bytes": 268976
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"sha256": "c1297071cf2cd8ef2da147b116cd01823b41f9af63bc2a603e075a72ec9936ca",
|
| 289 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0024.pt",
|
| 290 |
"bytes": 268976
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"sha256": "c95fcb3a196dfed881e2a01f5b54b8dea25fbfe13f5eae71336af1c671f71d14",
|
| 294 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0032.pt",
|
| 295 |
"bytes": 268976
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"sha256": "84d1bb51dafe0dc9ab687d2bfaa6f2e590a6a257a72053a9d05610754a27e554",
|
| 299 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0040.pt",
|
| 300 |
"bytes": 268976
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"sha256": "a2a4ffa22bf4208049e3b06accfb87699cb28b0987592bbbbdada6aea7410750",
|
| 304 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0048.pt",
|
| 305 |
"bytes": 268976
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"sha256": "99d7cae2f26351d8183ceb0b2c152adf512324e147ffbce0c8b0a5fd4c34163a",
|
| 309 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0056.pt",
|
| 310 |
"bytes": 268976
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"sha256": "e2a449c3025fe0223022be26a6a9c1ee8e6a60e46f146208f6a7a532f9dba37e",
|
| 314 |
"hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0064.pt",
|
| 315 |
"bytes": 268976
|
| 316 |
},
|