Backup runs/R1_aligned_v2/training.jsonl (67,583 bytes)
Browse files
byte_repair_v1/runs/R1_aligned_v2/training.jsonl
CHANGED
|
@@ -218,3 +218,13 @@
|
|
| 218 |
{"model": "R1", "step": 21800, "bytes_seen": 714342400, "loss": 2.4203907299041747, "throughput_bytes_per_second": 111046.04797711024, "gpu_hours": 1.7869023532052777, "lr": 0.0001648005915845802, "base_loss": 0.9858269095420837, "on_policy_rollout_loss": 3.1305906772613525, "rollout_steps": 2}
|
| 219 |
{"model": "R1", "step": 21900, "bytes_seen": 717619200, "loss": 2.419979512691498, "throughput_bytes_per_second": 111056.56614166145, "gpu_hours": 1.7949291391955555, "lr": 0.00016257459404780627, "base_loss": 1.005272626876831, "on_policy_rollout_loss": 3.141045570373535, "rollout_steps": 2}
|
| 220 |
{"model": "R1", "step": 22000, "bytes_seen": 720896000, "loss": 2.4284257745742797, "throughput_bytes_per_second": 111067.08895953855, "gpu_hours": 1.8029543293588888, "lr": 0.00016036693458142275, "base_loss": 1.0488934516906738, "on_policy_rollout_loss": 3.072699546813965, "rollout_steps": 2}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 218 |
{"model": "R1", "step": 21800, "bytes_seen": 714342400, "loss": 2.4203907299041747, "throughput_bytes_per_second": 111046.04797711024, "gpu_hours": 1.7869023532052777, "lr": 0.0001648005915845802, "base_loss": 0.9858269095420837, "on_policy_rollout_loss": 3.1305906772613525, "rollout_steps": 2}
|
| 219 |
{"model": "R1", "step": 21900, "bytes_seen": 717619200, "loss": 2.419979512691498, "throughput_bytes_per_second": 111056.56614166145, "gpu_hours": 1.7949291391955555, "lr": 0.00016257459404780627, "base_loss": 1.005272626876831, "on_policy_rollout_loss": 3.141045570373535, "rollout_steps": 2}
|
| 220 |
{"model": "R1", "step": 22000, "bytes_seen": 720896000, "loss": 2.4284257745742797, "throughput_bytes_per_second": 111067.08895953855, "gpu_hours": 1.8029543293588888, "lr": 0.00016036693458142275, "base_loss": 1.0488934516906738, "on_policy_rollout_loss": 3.072699546813965, "rollout_steps": 2}
|
| 221 |
+
{"model": "R1", "step": 22100, "bytes_seen": 724172800, "loss": 2.428906648159027, "throughput_bytes_per_second": 110978.95187602311, "gpu_hours": 1.8125879521355557, "lr": 0.00015817785498991859, "base_loss": 0.8135795593261719, "on_policy_rollout_loss": 3.1355481147766113, "rollout_steps": 2}
|
| 222 |
+
{"model": "R1", "step": 22200, "bytes_seen": 727449600, "loss": 2.427486517429352, "throughput_bytes_per_second": 110989.71965518885, "gpu_hours": 1.820613061832222, "lr": 0.00015600759504273253, "base_loss": 1.0093237161636353, "on_policy_rollout_loss": 3.141380786895752, "rollout_steps": 2}
|
| 223 |
+
{"model": "R1", "step": 22300, "bytes_seen": 730726400, "loss": 2.4360899806022642, "throughput_bytes_per_second": 111000.37688627254, "gpu_hours": 1.8286384357372223, "lr": 0.0001538563924479916, "base_loss": 1.1244903802871704, "on_policy_rollout_loss": 3.106019973754883, "rollout_steps": 2}
|
| 224 |
+
{"model": "R1", "step": 22400, "bytes_seen": 734003200, "loss": 2.4263689661026, "throughput_bytes_per_second": 111010.70582212046, "gpu_hours": 1.836667700361111, "lr": 0.00015172448282647517, "base_loss": 1.0551488399505615, "on_policy_rollout_loss": 3.1302640438079834, "rollout_steps": 2}
|
| 225 |
+
{"model": "R1", "step": 22500, "bytes_seen": 737280000, "loss": 2.426947093009949, "throughput_bytes_per_second": 111021.14913170812, "gpu_hours": 1.844693570565, "lr": 0.00014961209968580695, "base_loss": 0.8381801843643188, "on_policy_rollout_loss": 3.096601963043213, "rollout_steps": 2}
|
| 226 |
+
{"model": "R1", "step": 22600, "bytes_seen": 740556800, "loss": 2.427100417613983, "throughput_bytes_per_second": 111031.42914586794, "gpu_hours": 1.8527206558061111, "lr": 0.00014751947439487926, "base_loss": 0.9475950598716736, "on_policy_rollout_loss": 3.129026412963867, "rollout_steps": 2}
|
| 227 |
+
{"model": "R1", "step": 22700, "bytes_seen": 743833600, "loss": 2.433731286525726, "throughput_bytes_per_second": 111041.64086693428, "gpu_hours": 1.8607473991855554, "lr": 0.00014544683615851117, "base_loss": 1.0139217376708984, "on_policy_rollout_loss": 3.12241268157959, "rollout_steps": 2}
|
| 228 |
+
{"model": "R1", "step": 22800, "bytes_seen": 747110400, "loss": 2.423888545036316, "throughput_bytes_per_second": 111051.83524956417, "gpu_hours": 1.8687729581441666, "lr": 0.0001433944119923439, "base_loss": 1.0313037633895874, "on_policy_rollout_loss": 3.146143674850464, "rollout_steps": 2}
|
| 229 |
+
{"model": "R1", "step": 22900, "bytes_seen": 750387200, "loss": 2.4374578619003295, "throughput_bytes_per_second": 111061.8333234264, "gpu_hours": 1.8768003611275001, "lr": 0.00014136242669797558, "base_loss": 1.1203970909118652, "on_policy_rollout_loss": 3.1380374431610107, "rollout_steps": 2}
|
| 230 |
+
{"model": "R1", "step": 23000, "bytes_seen": 753664000, "loss": 2.4206141567230226, "throughput_bytes_per_second": 111071.61145060998, "gpu_hours": 1.8848300513241667, "lr": 0.00013935110283833885, "base_loss": 1.0707906484603882, "on_policy_rollout_loss": 3.1638894081115723, "rollout_steps": 2}
|