Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +133 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -121,3 +121,136 @@
|
|
| 121 |
{"event": "train", "epoch": 0, "step": 2400, "loss": 3.46912282705307, "perplexity": 32.10856532358906, "learning_rate": 0.0002994733325157695, "grad_norm": 0.6895716190338135, "tokens_per_sec": 113380.14105969662, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 122 |
{"event": "train", "epoch": 0, "step": 2420, "loss": 3.491629123687744, "perplexity": 32.83940360157507, "learning_rate": 0.00029946452339827035, "grad_norm": 0.5877766609191895, "tokens_per_sec": 113186.67170015186, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 123 |
{"event": "train", "epoch": 0, "step": 2440, "loss": 2.815937638282776, "perplexity": 16.708835281330597, "learning_rate": 0.00029945564135094795, "grad_norm": 0.8016887307167053, "tokens_per_sec": 113890.12094453946, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
{"event": "train", "epoch": 0, "step": 2400, "loss": 3.46912282705307, "perplexity": 32.10856532358906, "learning_rate": 0.0002994733325157695, "grad_norm": 0.6895716190338135, "tokens_per_sec": 113380.14105969662, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 122 |
{"event": "train", "epoch": 0, "step": 2420, "loss": 3.491629123687744, "perplexity": 32.83940360157507, "learning_rate": 0.00029946452339827035, "grad_norm": 0.5877766609191895, "tokens_per_sec": 113186.67170015186, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 123 |
{"event": "train", "epoch": 0, "step": 2440, "loss": 2.815937638282776, "perplexity": 16.708835281330597, "learning_rate": 0.00029945564135094795, "grad_norm": 0.8016887307167053, "tokens_per_sec": 113890.12094453946, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 124 |
+
{"event": "train", "epoch": 0, "step": 2460, "loss": 3.079291820526123, "perplexity": 21.74299899706504, "learning_rate": 0.00029944668637813616, "grad_norm": 0.6178374886512756, "tokens_per_sec": 114554.7994125521, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 125 |
+
{"event": "train", "epoch": 0, "step": 2480, "loss": 2.7583659291267395, "perplexity": 15.774045967883911, "learning_rate": 0.0002994376584842043, "grad_norm": 0.5049437880516052, "tokens_per_sec": 114421.37713969355, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 126 |
+
{"event": "train", "epoch": 0, "step": 2500, "loss": 3.2988815903663635, "perplexity": 27.082332834564045, "learning_rate": 0.00029942855767355736, "grad_norm": 0.678131639957428, "tokens_per_sec": 101988.7009194191, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 127 |
+
{"event": "train", "epoch": 0, "step": 2520, "loss": 2.934183955192566, "perplexity": 18.806150210472314, "learning_rate": 0.00029941938395063635, "grad_norm": 0.5913030505180359, "tokens_per_sec": 99260.63371001721, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 128 |
+
{"event": "train", "epoch": 0, "step": 2540, "loss": 3.284745156764984, "perplexity": 26.7021785781548, "learning_rate": 0.0002994101373199175, "grad_norm": 0.641316294670105, "tokens_per_sec": 101533.99462332626, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 129 |
+
{"event": "train", "epoch": 0, "step": 2560, "loss": 3.1843175888061523, "perplexity": 24.150802003604507, "learning_rate": 0.00029940081778591235, "grad_norm": 0.5363975167274475, "tokens_per_sec": 135048.46066968588, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 130 |
+
{"event": "train", "epoch": 0, "step": 2580, "loss": 2.725451409816742, "perplexity": 15.263302376513947, "learning_rate": 0.0002993914253531683, "grad_norm": 0.5075777769088745, "tokens_per_sec": 112212.83490773232, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 131 |
+
{"event": "train", "epoch": 0, "step": 2600, "loss": 3.140632927417755, "perplexity": 23.118494558160933, "learning_rate": 0.0002993819600262686, "grad_norm": 0.6176459789276123, "tokens_per_sec": 112543.25912169077, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 132 |
+
{"event": "train", "epoch": 0, "step": 2620, "loss": 3.4340530037879944, "perplexity": 31.002039845497762, "learning_rate": 0.00029937242180983154, "grad_norm": 0.6236518025398254, "tokens_per_sec": 134669.7396162706, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 133 |
+
{"event": "train", "epoch": 0, "step": 2640, "loss": 2.7741673588752747, "perplexity": 16.025278133412012, "learning_rate": 0.0002993628107085111, "grad_norm": 0.6058216691017151, "tokens_per_sec": 133747.65202379526, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 134 |
+
{"event": "train", "epoch": 0, "step": 2660, "loss": 2.6619825065135956, "perplexity": 14.324659698092887, "learning_rate": 0.0002993531267269972, "grad_norm": 0.7889980673789978, "tokens_per_sec": 134934.1704656462, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 135 |
+
{"event": "train", "epoch": 0, "step": 2680, "loss": 2.9527056217193604, "perplexity": 19.157717206225616, "learning_rate": 0.00029934336987001475, "grad_norm": 0.5849175453186035, "tokens_per_sec": 134877.49496511745, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 136 |
+
{"event": "train", "epoch": 0, "step": 2700, "loss": 3.065339207649231, "perplexity": 21.441733953928477, "learning_rate": 0.00029933354014232474, "grad_norm": 1.0489201545715332, "tokens_per_sec": 134372.9655177808, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 137 |
+
{"event": "train", "epoch": 0, "step": 2720, "loss": 2.8135774731636047, "perplexity": 16.669446171803173, "learning_rate": 0.0002993236375487234, "grad_norm": 0.6522538065910339, "tokens_per_sec": 134470.1221159246, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 138 |
+
{"event": "train", "epoch": 0, "step": 2740, "loss": 3.415963351726532, "perplexity": 30.446265767017945, "learning_rate": 0.00029931366209404257, "grad_norm": 1.276015043258667, "tokens_per_sec": 135415.3071223634, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 139 |
+
{"event": "train", "epoch": 0, "step": 2760, "loss": 3.3785842657089233, "perplexity": 29.329219325554565, "learning_rate": 0.00029930361378314974, "grad_norm": 0.6388046741485596, "tokens_per_sec": 135297.9979524012, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 140 |
+
{"event": "train", "epoch": 0, "step": 2780, "loss": 3.250254213809967, "perplexity": 25.796897011180366, "learning_rate": 0.000299293492620948, "grad_norm": 0.6032503843307495, "tokens_per_sec": 135109.39724470678, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 141 |
+
{"event": "train", "epoch": 0, "step": 2800, "loss": 3.0326388478279114, "perplexity": 20.75192156285558, "learning_rate": 0.0002992832986123757, "grad_norm": 0.8859052062034607, "tokens_per_sec": 134803.14752837762, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 142 |
+
{"event": "train", "epoch": 0, "step": 2820, "loss": 3.102148115634918, "perplexity": 22.24568629946602, "learning_rate": 0.000299273031762407, "grad_norm": 0.6345592737197876, "tokens_per_sec": 135350.6286229761, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 143 |
+
{"event": "train", "epoch": 0, "step": 2840, "loss": 3.0747233033180237, "perplexity": 21.64389228971441, "learning_rate": 0.0002992626920760514, "grad_norm": 0.5578694343566895, "tokens_per_sec": 134289.85634080804, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 144 |
+
{"event": "train", "epoch": 0, "step": 2860, "loss": 2.525826096534729, "perplexity": 12.5012182121887, "learning_rate": 0.00029925227955835424, "grad_norm": 0.592522382736206, "tokens_per_sec": 134983.07146968064, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 145 |
+
{"event": "train", "epoch": 0, "step": 2880, "loss": 3.2054648995399475, "perplexity": 24.666965011900132, "learning_rate": 0.00029924179421439613, "grad_norm": 1.24861478805542, "tokens_per_sec": 134471.96405704535, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 146 |
+
{"event": "train", "epoch": 0, "step": 2900, "loss": 2.554780900478363, "perplexity": 12.868479870842572, "learning_rate": 0.0002992312360492934, "grad_norm": 0.6011277437210083, "tokens_per_sec": 134409.36635437244, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 147 |
+
{"event": "train", "epoch": 0, "step": 2920, "loss": 3.129566252231598, "perplexity": 22.864060156271126, "learning_rate": 0.00029922060506819764, "grad_norm": 0.6538704037666321, "tokens_per_sec": 135342.36489737476, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 148 |
+
{"event": "train", "epoch": 0, "step": 2940, "loss": 3.2179402709007263, "perplexity": 24.97662208816737, "learning_rate": 0.00029920990127629615, "grad_norm": 0.6825475096702576, "tokens_per_sec": 135385.02722398491, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 149 |
+
{"event": "train", "epoch": 0, "step": 2960, "loss": 2.9519062638282776, "perplexity": 19.142409452803534, "learning_rate": 0.00029919912467881176, "grad_norm": 0.517304003238678, "tokens_per_sec": 134928.20935237108, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 150 |
+
{"event": "train", "epoch": 0, "step": 2980, "loss": 3.388947606086731, "perplexity": 29.634748425016586, "learning_rate": 0.00029918827528100294, "grad_norm": 0.5365781188011169, "tokens_per_sec": 134986.385827893, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 151 |
+
{"event": "train", "epoch": 0, "step": 3000, "loss": 3.7732375264167786, "perplexity": 43.52073653301533, "learning_rate": 0.0002991773530881634, "grad_norm": 0.9687048196792603, "tokens_per_sec": 134924.36802511592, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 152 |
+
{"event": "train", "epoch": 0, "step": 3020, "loss": 2.626443088054657, "perplexity": 13.824509783125961, "learning_rate": 0.00029916635810562243, "grad_norm": 0.5325588583946228, "tokens_per_sec": 129700.8698744221, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 153 |
+
{"event": "train", "epoch": 0, "step": 3040, "loss": 3.0667461156845093, "perplexity": 21.471921732453197, "learning_rate": 0.00029915529033874513, "grad_norm": 0.5942113995552063, "tokens_per_sec": 134945.03429322687, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 154 |
+
{"event": "train", "epoch": 0, "step": 3060, "loss": 2.998217523097992, "perplexity": 20.049766806731643, "learning_rate": 0.00029914414979293196, "grad_norm": 0.5828272700309753, "tokens_per_sec": 131351.34407930073, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 155 |
+
{"event": "train", "epoch": 0, "step": 3080, "loss": 3.2851202189922333, "perplexity": 26.71219543507964, "learning_rate": 0.0002991329364736187, "grad_norm": 0.5895505547523499, "tokens_per_sec": 134735.88220323864, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 156 |
+
{"event": "train", "epoch": 0, "step": 3100, "loss": 3.179067850112915, "perplexity": 24.024348817363936, "learning_rate": 0.0002991216503862767, "grad_norm": 1.4114855527877808, "tokens_per_sec": 135232.50000196788, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 157 |
+
{"event": "train", "epoch": 0, "step": 3120, "loss": 2.9001834392547607, "perplexity": 18.17747952692388, "learning_rate": 0.0002991102915364131, "grad_norm": 0.6633157134056091, "tokens_per_sec": 134421.59199645557, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 158 |
+
{"event": "train", "epoch": 0, "step": 3140, "loss": 3.015059232711792, "perplexity": 20.390298677465008, "learning_rate": 0.00029909885992957023, "grad_norm": 0.6890828609466553, "tokens_per_sec": 134819.01554295377, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 159 |
+
{"event": "train", "epoch": 0, "step": 3160, "loss": 2.9649243354797363, "perplexity": 19.393235806072024, "learning_rate": 0.0002990873555713263, "grad_norm": 0.5076770186424255, "tokens_per_sec": 134493.0183970314, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 160 |
+
{"event": "train", "epoch": 0, "step": 3180, "loss": 2.565821886062622, "perplexity": 13.011347821743694, "learning_rate": 0.00029907577846729447, "grad_norm": 0.566415011882782, "tokens_per_sec": 134902.11912191243, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 161 |
+
{"event": "train", "epoch": 0, "step": 3200, "loss": 2.8830790519714355, "perplexity": 17.86920878395431, "learning_rate": 0.0002990641286231237, "grad_norm": 0.5408375263214111, "tokens_per_sec": 134902.51635941037, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 162 |
+
{"event": "train", "epoch": 0, "step": 3220, "loss": 2.6359197199344635, "perplexity": 13.956142304613776, "learning_rate": 0.00029905240604449854, "grad_norm": 0.5891720056533813, "tokens_per_sec": 134938.14483379526, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 163 |
+
{"event": "train", "epoch": 0, "step": 3240, "loss": 2.78685599565506, "perplexity": 16.229912587541925, "learning_rate": 0.00029904061073713863, "grad_norm": 0.5307798981666565, "tokens_per_sec": 134815.44491397406, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 164 |
+
{"event": "train", "epoch": 0, "step": 3260, "loss": 3.2240028977394104, "perplexity": 25.12850597001498, "learning_rate": 0.00029902874270679966, "grad_norm": 0.6424093246459961, "tokens_per_sec": 134442.4990531044, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 165 |
+
{"event": "train", "epoch": 0, "step": 3280, "loss": 3.499016046524048, "perplexity": 33.08288391997488, "learning_rate": 0.00029901680195927247, "grad_norm": 0.5891613364219666, "tokens_per_sec": 134844.6767362709, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 166 |
+
{"event": "train", "epoch": 0, "step": 3300, "loss": 3.172267258167267, "perplexity": 23.861523306754847, "learning_rate": 0.0002990047885003836, "grad_norm": 0.5834274888038635, "tokens_per_sec": 132084.76542089845, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 167 |
+
{"event": "train", "epoch": 0, "step": 3320, "loss": 2.8185799717903137, "perplexity": 16.753043978004843, "learning_rate": 0.0002989927023359945, "grad_norm": 0.5318182706832886, "tokens_per_sec": 134980.02240386873, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 168 |
+
{"event": "train", "epoch": 0, "step": 3340, "loss": 2.418489098548889, "perplexity": 11.228880759415969, "learning_rate": 0.0002989805434720027, "grad_norm": 0.5350157022476196, "tokens_per_sec": 133790.35665564393, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 169 |
+
{"event": "train", "epoch": 0, "step": 3360, "loss": 3.5777627825737, "perplexity": 35.79337364550415, "learning_rate": 0.00029896831191434113, "grad_norm": 0.6387715339660645, "tokens_per_sec": 135555.1436408235, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 170 |
+
{"event": "train", "epoch": 0, "step": 3380, "loss": 3.141819477081299, "perplexity": 23.14594208080465, "learning_rate": 0.00029895600766897804, "grad_norm": 0.6195865869522095, "tokens_per_sec": 134630.69187027786, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 171 |
+
{"event": "train", "epoch": 0, "step": 3400, "loss": 2.722721219062805, "perplexity": 15.221687483638503, "learning_rate": 0.00029894363074191706, "grad_norm": 0.7536229491233826, "tokens_per_sec": 135329.97120080664, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 172 |
+
{"event": "train", "epoch": 0, "step": 3420, "loss": 2.8741933703422546, "perplexity": 17.711132033655034, "learning_rate": 0.0002989311811391973, "grad_norm": 0.6474753618240356, "tokens_per_sec": 134328.5753106812, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 173 |
+
{"event": "train", "epoch": 0, "step": 3440, "loss": 2.8782312273979187, "perplexity": 17.782791631343784, "learning_rate": 0.0002989186588668936, "grad_norm": 0.5218234658241272, "tokens_per_sec": 134853.14442760075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 174 |
+
{"event": "train", "epoch": 0, "step": 3460, "loss": 3.018690288066864, "perplexity": 20.464471562099575, "learning_rate": 0.00029890606393111625, "grad_norm": 0.5405023097991943, "tokens_per_sec": 134224.8062612677, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 175 |
+
{"event": "train", "epoch": 0, "step": 3480, "loss": 2.4134246706962585, "perplexity": 11.172156661734503, "learning_rate": 0.0002988933963380105, "grad_norm": 0.5797339677810669, "tokens_per_sec": 134122.89891912357, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 176 |
+
{"event": "train", "epoch": 0, "step": 3500, "loss": 2.4685431122779846, "perplexity": 11.805235414593982, "learning_rate": 0.0002988806560937576, "grad_norm": 0.5119703412055969, "tokens_per_sec": 135225.98030431982, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 177 |
+
{"event": "train", "epoch": 0, "step": 3520, "loss": 2.92650043964386, "perplexity": 18.662206567686567, "learning_rate": 0.00029886784320457395, "grad_norm": 0.5680127739906311, "tokens_per_sec": 134620.27417139924, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 178 |
+
{"event": "train", "epoch": 0, "step": 3540, "loss": 2.868202269077301, "perplexity": 17.605340069743832, "learning_rate": 0.00029885495767671135, "grad_norm": 0.6858769655227661, "tokens_per_sec": 135220.9246450473, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 179 |
+
{"event": "train", "epoch": 0, "step": 3560, "loss": 2.536550283432007, "perplexity": 12.636005060620048, "learning_rate": 0.0002988419995164572, "grad_norm": 0.5043861269950867, "tokens_per_sec": 134618.16445403025, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 180 |
+
{"event": "train", "epoch": 0, "step": 3580, "loss": 2.9934170246124268, "perplexity": 19.953748583600568, "learning_rate": 0.00029882896873013464, "grad_norm": 0.6571323871612549, "tokens_per_sec": 135354.2276576174, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 181 |
+
{"event": "train", "epoch": 0, "step": 3600, "loss": 3.5292473435401917, "perplexity": 34.098293653143365, "learning_rate": 0.0002988158653241018, "grad_norm": 0.6039084792137146, "tokens_per_sec": 134600.76180850583, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 182 |
+
{"event": "train", "epoch": 0, "step": 3620, "loss": 2.618943989276886, "perplexity": 13.721226169742621, "learning_rate": 0.0002988026893047522, "grad_norm": 0.520337700843811, "tokens_per_sec": 134219.95626099262, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 183 |
+
{"event": "train", "epoch": 0, "step": 3640, "loss": 2.7431410551071167, "perplexity": 15.535707049081395, "learning_rate": 0.00029878944067851493, "grad_norm": 0.5780593156814575, "tokens_per_sec": 134551.74218812716, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 184 |
+
{"event": "train", "epoch": 0, "step": 3660, "loss": 2.6732718348503113, "perplexity": 14.487291760843691, "learning_rate": 0.0002987761194518546, "grad_norm": 0.6220414042472839, "tokens_per_sec": 135121.21919907822, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 185 |
+
{"event": "train", "epoch": 0, "step": 3680, "loss": 3.028438091278076, "perplexity": 20.664930634208094, "learning_rate": 0.0002987627256312712, "grad_norm": 0.56772381067276, "tokens_per_sec": 134790.7201923803, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 186 |
+
{"event": "train", "epoch": 0, "step": 3700, "loss": 2.9331496357917786, "perplexity": 18.786708700554183, "learning_rate": 0.0002987492592233002, "grad_norm": 0.6119143962860107, "tokens_per_sec": 134495.91388438197, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 187 |
+
{"event": "train", "epoch": 0, "step": 3720, "loss": 2.7589868307113647, "perplexity": 15.783843139246185, "learning_rate": 0.0002987357202345124, "grad_norm": 0.5927655100822449, "tokens_per_sec": 134218.64551027204, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 188 |
+
{"event": "train", "epoch": 0, "step": 3740, "loss": 2.5039377212524414, "perplexity": 12.230559798908933, "learning_rate": 0.00029872210867151386, "grad_norm": 0.4580960273742676, "tokens_per_sec": 133678.44513199193, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 189 |
+
{"event": "train", "epoch": 0, "step": 3760, "loss": 2.7410839200019836, "perplexity": 15.50378085023866, "learning_rate": 0.0002987084245409463, "grad_norm": 0.5746908783912659, "tokens_per_sec": 134383.60776583594, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 190 |
+
{"event": "train", "epoch": 0, "step": 3780, "loss": 2.1405538618564606, "perplexity": 8.50414644708715, "learning_rate": 0.00029869466784948677, "grad_norm": 0.5112942457199097, "tokens_per_sec": 134721.4863704222, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 191 |
+
{"event": "train", "epoch": 0, "step": 3800, "loss": 2.614384412765503, "perplexity": 13.658805603008872, "learning_rate": 0.0002986808386038478, "grad_norm": 0.5415308475494385, "tokens_per_sec": 134802.75087587145, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 192 |
+
{"event": "train", "epoch": 0, "step": 3820, "loss": 2.1849746704101562, "perplexity": 8.890423359742112, "learning_rate": 0.0002986669368107771, "grad_norm": 0.48591524362564087, "tokens_per_sec": 134739.84489898357, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 193 |
+
{"event": "train", "epoch": 0, "step": 3840, "loss": 2.638093888759613, "perplexity": 13.986518323462779, "learning_rate": 0.0002986529624770579, "grad_norm": 0.5242927074432373, "tokens_per_sec": 134511.31470739358, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 194 |
+
{"event": "train", "epoch": 0, "step": 3860, "loss": 3.1930496096611023, "perplexity": 24.362610723609706, "learning_rate": 0.0002986389156095091, "grad_norm": 0.5864602327346802, "tokens_per_sec": 135252.4621685827, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 195 |
+
{"event": "train", "epoch": 0, "step": 3880, "loss": 2.603643774986267, "perplexity": 13.51288635520627, "learning_rate": 0.0002986247962149845, "grad_norm": 0.5625761151313782, "tokens_per_sec": 134890.60025203848, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 196 |
+
{"event": "train", "epoch": 0, "step": 3900, "loss": 2.407840132713318, "perplexity": 11.109939218042635, "learning_rate": 0.00029861060430037373, "grad_norm": 0.4464544653892517, "tokens_per_sec": 135039.96843279147, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 197 |
+
{"event": "train", "epoch": 0, "step": 3920, "loss": 3.254286050796509, "perplexity": 25.90111585019374, "learning_rate": 0.0002985963398726013, "grad_norm": 0.6770306825637817, "tokens_per_sec": 135421.57822123822, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 198 |
+
{"event": "train", "epoch": 0, "step": 3940, "loss": 3.095300793647766, "perplexity": 22.093883237886136, "learning_rate": 0.0002985820029386277, "grad_norm": 0.5516052842140198, "tokens_per_sec": 134884.77536567557, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 199 |
+
{"event": "train", "epoch": 0, "step": 3960, "loss": 2.6268062591552734, "perplexity": 13.82953135734951, "learning_rate": 0.0002985675935054482, "grad_norm": 0.4638791084289551, "tokens_per_sec": 135284.68022411125, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 200 |
+
{"event": "train", "epoch": 0, "step": 3980, "loss": 2.7577452659606934, "perplexity": 15.764258636204453, "learning_rate": 0.00029855311158009414, "grad_norm": 0.7153812646865845, "tokens_per_sec": 135098.63973585577, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 201 |
+
{"event": "train", "epoch": 0, "step": 4000, "loss": 2.8402316570281982, "perplexity": 17.11973098381961, "learning_rate": 0.0002985385571696315, "grad_norm": 0.5645555257797241, "tokens_per_sec": 134309.93491821046, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 202 |
+
{"event": "train", "epoch": 0, "step": 4020, "loss": 2.4944668412208557, "perplexity": 12.115272432490999, "learning_rate": 0.00029852393028116203, "grad_norm": 0.49216729402542114, "tokens_per_sec": 134953.51459372733, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 203 |
+
{"event": "train", "epoch": 0, "step": 4040, "loss": 2.4164453744888306, "perplexity": 11.205955460112852, "learning_rate": 0.000298509230921823, "grad_norm": 0.5198336839675903, "tokens_per_sec": 134405.55451221386, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 204 |
+
{"event": "train", "epoch": 0, "step": 4060, "loss": 2.861155092716217, "perplexity": 17.481708272571673, "learning_rate": 0.0002984944590987867, "grad_norm": 0.859401524066925, "tokens_per_sec": 134433.81986278726, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 205 |
+
{"event": "train", "epoch": 0, "step": 4080, "loss": 2.9998698234558105, "perplexity": 20.08292242757951, "learning_rate": 0.00029847961481926076, "grad_norm": 0.6424986720085144, "tokens_per_sec": 134572.82151928233, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 206 |
+
{"event": "train", "epoch": 0, "step": 4100, "loss": 2.824454367160797, "perplexity": 16.851747610227964, "learning_rate": 0.0002984646980904885, "grad_norm": 0.5019730925559998, "tokens_per_sec": 134824.17345207054, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 207 |
+
{"event": "train", "epoch": 0, "step": 4120, "loss": 2.5642709136009216, "perplexity": 12.991183220994984, "learning_rate": 0.0002984497089197483, "grad_norm": 0.5825138688087463, "tokens_per_sec": 134713.16724594554, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 208 |
+
{"event": "train", "epoch": 0, "step": 4140, "loss": 2.9704548120498657, "perplexity": 19.50078677223485, "learning_rate": 0.00029843464731435396, "grad_norm": 0.579096257686615, "tokens_per_sec": 135311.85117158273, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 209 |
+
{"event": "train", "epoch": 0, "step": 4160, "loss": 2.656635344028473, "perplexity": 14.248267837007166, "learning_rate": 0.0002984195132816548, "grad_norm": 0.5055664777755737, "tokens_per_sec": 134579.2783855847, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 210 |
+
{"event": "train", "epoch": 0, "step": 4180, "loss": 3.0395853519439697, "perplexity": 20.896576713526816, "learning_rate": 0.0002984043068290351, "grad_norm": 0.7078943848609924, "tokens_per_sec": 134510.91977043712, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 211 |
+
{"event": "train", "epoch": 0, "step": 4200, "loss": 2.4773158729076385, "perplexity": 11.90925552369554, "learning_rate": 0.000298389027963915, "grad_norm": 0.5031247735023499, "tokens_per_sec": 135198.84382860386, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 212 |
+
{"event": "train", "epoch": 0, "step": 4220, "loss": 2.6285228729248047, "perplexity": 13.853291709142551, "learning_rate": 0.00029837367669374964, "grad_norm": 0.5482646226882935, "tokens_per_sec": 134341.44282478886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 213 |
+
{"event": "train", "epoch": 0, "step": 4240, "loss": 2.445284426212311, "perplexity": 11.533829655832092, "learning_rate": 0.0002983582530260292, "grad_norm": 0.5156395435333252, "tokens_per_sec": 134837.40066359658, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 214 |
+
{"event": "train", "epoch": 0, "step": 4260, "loss": 2.175890773534775, "perplexity": 8.810029368947662, "learning_rate": 0.0002983427569682799, "grad_norm": 0.5821630954742432, "tokens_per_sec": 134555.4305943628, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 215 |
+
{"event": "train", "epoch": 0, "step": 4280, "loss": 2.820159435272217, "perplexity": 16.779525707135527, "learning_rate": 0.0002983271885280625, "grad_norm": 0.4684588313102722, "tokens_per_sec": 134282.11522517144, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 216 |
+
{"event": "train", "epoch": 0, "step": 4300, "loss": 2.9331604838371277, "perplexity": 18.786912500727542, "learning_rate": 0.0002983115477129738, "grad_norm": 0.6429683566093445, "tokens_per_sec": 134682.93654155027, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 217 |
+
{"event": "train", "epoch": 0, "step": 4320, "loss": 2.387382924556732, "perplexity": 10.884969847179123, "learning_rate": 0.0002982958345306455, "grad_norm": 0.5309492945671082, "tokens_per_sec": 133857.8542960186, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 218 |
+
{"event": "train", "epoch": 0, "step": 4340, "loss": 2.5947477221488953, "perplexity": 13.393208125407293, "learning_rate": 0.00029828004898874475, "grad_norm": 0.4991573095321655, "tokens_per_sec": 132021.9604142432, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 219 |
+
{"event": "train", "epoch": 0, "step": 4360, "loss": 2.7083404064178467, "perplexity": 15.004353711439094, "learning_rate": 0.0002982641910949738, "grad_norm": 0.4825790822505951, "tokens_per_sec": 134818.222053502, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 220 |
+
{"event": "train", "epoch": 0, "step": 4380, "loss": 2.8929991722106934, "perplexity": 18.047355641864634, "learning_rate": 0.00029824826085707046, "grad_norm": 0.5252625346183777, "tokens_per_sec": 134258.11032224537, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 221 |
+
{"event": "train", "epoch": 0, "step": 4400, "loss": 2.5632715821266174, "perplexity": 12.978207207463047, "learning_rate": 0.0002982322582828077, "grad_norm": 0.5881826877593994, "tokens_per_sec": 135370.89234957553, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 222 |
+
{"event": "train", "epoch": 0, "step": 4420, "loss": 2.3064430356025696, "perplexity": 10.03865394048437, "learning_rate": 0.0002982161833799938, "grad_norm": 0.5300384759902954, "tokens_per_sec": 134730.73104721983, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 223 |
+
{"event": "train", "epoch": 0, "step": 4440, "loss": 2.2710273265838623, "perplexity": 9.689349829413953, "learning_rate": 0.00029820003615647266, "grad_norm": 0.46173202991485596, "tokens_per_sec": 134308.09741320089, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 224 |
+
{"event": "train", "epoch": 0, "step": 4460, "loss": 3.231900990009308, "perplexity": 25.327759052605945, "learning_rate": 0.0002981838166201226, "grad_norm": 0.6127412915229797, "tokens_per_sec": 134490.25463903908, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 225 |
+
{"event": "train", "epoch": 0, "step": 4480, "loss": 2.6709596514701843, "perplexity": 14.453833181702855, "learning_rate": 0.0002981675247788583, "grad_norm": 0.6796795129776001, "tokens_per_sec": 134904.89983352702, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 226 |
+
{"event": "train", "epoch": 0, "step": 4500, "loss": 2.839854419231415, "perplexity": 17.11327399220897, "learning_rate": 0.00029815116064062913, "grad_norm": 0.6157224178314209, "tokens_per_sec": 135126.79881743487, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 227 |
+
{"event": "train", "epoch": 0, "step": 4520, "loss": 2.4502640068531036, "perplexity": 11.59140652606062, "learning_rate": 0.00029813472421341973, "grad_norm": 0.5412400960922241, "tokens_per_sec": 134898.6764949241, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 228 |
+
{"event": "train", "epoch": 0, "step": 4540, "loss": 2.4941898584365845, "perplexity": 12.111917175296133, "learning_rate": 0.0002981182155052501, "grad_norm": 0.4929715394973755, "tokens_per_sec": 134621.592778335, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 229 |
+
{"event": "train", "epoch": 0, "step": 4560, "loss": 2.7365055680274963, "perplexity": 15.432961326629595, "learning_rate": 0.0002981016345241755, "grad_norm": 0.474637508392334, "tokens_per_sec": 133976.85749740945, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 230 |
+
{"event": "train", "epoch": 0, "step": 4580, "loss": 3.0124269127845764, "perplexity": 20.336695469282834, "learning_rate": 0.00029808498127828644, "grad_norm": 0.5065128207206726, "tokens_per_sec": 134323.98036728273, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 231 |
+
{"event": "train", "epoch": 0, "step": 4600, "loss": 3.5242682099342346, "perplexity": 33.92893567120907, "learning_rate": 0.0002980682557757089, "grad_norm": 0.5240969657897949, "tokens_per_sec": 134641.5068834819, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 232 |
+
{"event": "train", "epoch": 0, "step": 4620, "loss": 2.6563519835472107, "perplexity": 14.244231012940935, "learning_rate": 0.0002980514580246037, "grad_norm": 0.5339817404747009, "tokens_per_sec": 134799.44552909478, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 233 |
+
{"event": "train", "epoch": 0, "step": 4640, "loss": 2.4657957553863525, "perplexity": 11.772846731728297, "learning_rate": 0.0002980345880331671, "grad_norm": 0.4744551479816437, "tokens_per_sec": 133726.8303872795, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 234 |
+
{"event": "train", "epoch": 0, "step": 4660, "loss": 2.733088970184326, "perplexity": 15.380323077316305, "learning_rate": 0.00029801764580963085, "grad_norm": 0.5393480658531189, "tokens_per_sec": 135242.48034865747, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 235 |
+
{"event": "train", "epoch": 0, "step": 4680, "loss": 2.210896611213684, "perplexity": 9.123893311789798, "learning_rate": 0.00029800063136226177, "grad_norm": 0.45320650935173035, "tokens_per_sec": 133462.69927257227, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 236 |
+
{"event": "train", "epoch": 0, "step": 4700, "loss": 2.167042165994644, "perplexity": 8.73241676478971, "learning_rate": 0.0002979835446993616, "grad_norm": 0.5206497311592102, "tokens_per_sec": 134826.15736835857, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 237 |
+
{"event": "train", "epoch": 0, "step": 4720, "loss": 2.700983762741089, "perplexity": 14.894377053124161, "learning_rate": 0.000297966385829268, "grad_norm": 0.5028563737869263, "tokens_per_sec": 134510.39319143604, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 238 |
+
{"event": "train", "epoch": 0, "step": 4740, "loss": 2.8331199288368225, "perplexity": 16.998412015442426, "learning_rate": 0.0002979491547603535, "grad_norm": 0.5656957030296326, "tokens_per_sec": 133628.66569827928, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 239 |
+
{"event": "train", "epoch": 0, "step": 4760, "loss": 2.5934383869171143, "perplexity": 13.37568340151497, "learning_rate": 0.00029793185150102543, "grad_norm": 0.533761203289032, "tokens_per_sec": 134571.76739586372, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 240 |
+
{"event": "train", "epoch": 0, "step": 4780, "loss": 2.5082996487617493, "perplexity": 12.284025135339817, "learning_rate": 0.000297914476059727, "grad_norm": 0.48433956503868103, "tokens_per_sec": 134317.67924177952, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 241 |
+
{"event": "train", "epoch": 0, "step": 4800, "loss": 2.7794533371925354, "perplexity": 16.11021168684962, "learning_rate": 0.00029789702844493644, "grad_norm": 0.5578612089157104, "tokens_per_sec": 134947.15426843337, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 242 |
+
{"event": "train", "epoch": 0, "step": 4820, "loss": 3.248491644859314, "perplexity": 25.751468248909124, "learning_rate": 0.0002978795086651672, "grad_norm": 0.8942722082138062, "tokens_per_sec": 134117.40189662811, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 243 |
+
{"event": "train", "epoch": 0, "step": 4840, "loss": 2.965055465698242, "perplexity": 19.395779012062746, "learning_rate": 0.00029786191672896765, "grad_norm": 0.5161387324333191, "tokens_per_sec": 135057.4848417848, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 244 |
+
{"event": "train", "epoch": 0, "step": 4860, "loss": 2.7326087951660156, "perplexity": 15.372939603222802, "learning_rate": 0.0002978442526449218, "grad_norm": 0.5121952295303345, "tokens_per_sec": 134423.43260829806, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 245 |
+
{"event": "train", "epoch": 0, "step": 4880, "loss": 2.1793544590473175, "perplexity": 8.840597448584523, "learning_rate": 0.0002978265164216484, "grad_norm": 0.5649278163909912, "tokens_per_sec": 134427.24546461087, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 246 |
+
{"event": "train", "epoch": 0, "step": 4900, "loss": 2.5178991556167603, "perplexity": 12.40251352398723, "learning_rate": 0.0002978087080678021, "grad_norm": 0.5011149048805237, "tokens_per_sec": 134862.67185161854, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 247 |
+
{"event": "train", "epoch": 0, "step": 4920, "loss": 2.4468407034873962, "perplexity": 11.551793467530436, "learning_rate": 0.00029779082759207205, "grad_norm": 0.5044282674789429, "tokens_per_sec": 135086.02508703404, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 248 |
+
{"event": "train", "epoch": 0, "step": 4940, "loss": 2.611166477203369, "perplexity": 13.614923090139383, "learning_rate": 0.00029777287500318294, "grad_norm": 0.5168914198875427, "tokens_per_sec": 134627.6587029647, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 249 |
+
{"event": "train", "epoch": 0, "step": 4960, "loss": 2.4364976286888123, "perplexity": 11.432928179594205, "learning_rate": 0.00029775485030989455, "grad_norm": 0.5031206607818604, "tokens_per_sec": 135182.75331910414, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 250 |
+
{"event": "train", "epoch": 0, "step": 4980, "loss": 2.173914909362793, "perplexity": 8.792639133592619, "learning_rate": 0.0002977367535210019, "grad_norm": 0.41913357377052307, "tokens_per_sec": 134594.1710722176, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 251 |
+
{"event": "train", "epoch": 0, "step": 5000, "loss": 2.9597201347351074, "perplexity": 19.292571679207654, "learning_rate": 0.0002977185846453351, "grad_norm": 0.6947238445281982, "tokens_per_sec": 134877.230238088, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 252 |
+
{"event": "train", "epoch": 0, "step": 5020, "loss": 2.627877414226532, "perplexity": 13.844352866646336, "learning_rate": 0.00029770034369175943, "grad_norm": 0.5588812828063965, "tokens_per_sec": 134936.95249876785, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 253 |
+
{"event": "train", "epoch": 0, "step": 5040, "loss": 2.785085141658783, "perplexity": 16.201197214847397, "learning_rate": 0.00029768203066917565, "grad_norm": 0.5511414408683777, "tokens_per_sec": 133377.08764486332, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 254 |
+
{"event": "train", "epoch": 0, "step": 5060, "loss": 2.959497392177582, "perplexity": 19.288274881008256, "learning_rate": 0.0002976636455865192, "grad_norm": 0.576752245426178, "tokens_per_sec": 135255.12423608417, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 255 |
+
{"event": "train", "epoch": 0, "step": 5080, "loss": 2.556455075740814, "perplexity": 12.890042005867912, "learning_rate": 0.000297645188452761, "grad_norm": 0.4632514715194702, "tokens_per_sec": 134223.6264992114, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 256 |
+
{"event": "train", "epoch": 0, "step": 5100, "loss": 2.8122610449790955, "perplexity": 16.647516480641194, "learning_rate": 0.0002976266592769068, "grad_norm": 0.6443740725517273, "tokens_per_sec": 134261.7826474312, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|