Backup runs/R1_aligned_v2/training.jsonl (49,914 bytes)
Browse files
byte_repair_v1/runs/R1_aligned_v2/training.jsonl
CHANGED
|
@@ -158,3 +158,13 @@
|
|
| 158 |
{"model": "R1", "step": 15800, "bytes_seen": 517734400, "loss": 2.441078269481659, "throughput_bytes_per_second": 110982.94370110887, "gpu_hours": 1.2958307494386112, "lr": 0.0003217783736669338, "base_loss": 1.7162209749221802, "on_policy_rollout_loss": 3.0977413654327393, "rollout_steps": 2}
|
| 159 |
{"model": "R1", "step": 15900, "bytes_seen": 521011200, "loss": 2.4275813412666323, "throughput_bytes_per_second": 110997.3174743745, "gpu_hours": 1.3038633421636112, "lr": 0.00031895444789773684, "base_loss": 0.891711950302124, "on_policy_rollout_loss": 3.0731377601623535, "rollout_steps": 2}
|
| 160 |
{"model": "R1", "step": 16000, "bytes_seen": 524288000, "loss": 2.4519691252708435, "throughput_bytes_per_second": 111011.61427690764, "gpu_hours": 1.3118947643827779, "lr": 0.0003161317319455972, "base_loss": 1.6553153991699219, "on_policy_rollout_loss": 3.09867262840271, "rollout_steps": 2}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
{"model": "R1", "step": 15800, "bytes_seen": 517734400, "loss": 2.441078269481659, "throughput_bytes_per_second": 110982.94370110887, "gpu_hours": 1.2958307494386112, "lr": 0.0003217783736669338, "base_loss": 1.7162209749221802, "on_policy_rollout_loss": 3.0977413654327393, "rollout_steps": 2}
|
| 159 |
{"model": "R1", "step": 15900, "bytes_seen": 521011200, "loss": 2.4275813412666323, "throughput_bytes_per_second": 110997.3174743745, "gpu_hours": 1.3038633421636112, "lr": 0.00031895444789773684, "base_loss": 0.891711950302124, "on_policy_rollout_loss": 3.0731377601623535, "rollout_steps": 2}
|
| 160 |
{"model": "R1", "step": 16000, "bytes_seen": 524288000, "loss": 2.4519691252708435, "throughput_bytes_per_second": 111011.61427690764, "gpu_hours": 1.3118947643827779, "lr": 0.0003161317319455972, "base_loss": 1.6553153991699219, "on_policy_rollout_loss": 3.09867262840271, "rollout_steps": 2}
|
| 161 |
+
{"model": "R1", "step": 16100, "bytes_seen": 527564800, "loss": 2.4404510354995725, "throughput_bytes_per_second": 110894.54813336326, "gpu_hours": 1.3214876677394445, "lr": 0.0003133105349820145, "base_loss": 1.0655542612075806, "on_policy_rollout_loss": 3.194819450378418, "rollout_steps": 2}
|
| 162 |
+
{"model": "R1", "step": 16200, "bytes_seen": 530841600, "loss": 2.415243809223175, "throughput_bytes_per_second": 110909.5979072323, "gpu_hours": 1.3295152338694445, "lr": 0.0003104911660121137, "base_loss": 1.0539488792419434, "on_policy_rollout_loss": 3.1178693771362305, "rollout_steps": 2}
|
| 163 |
+
{"model": "R1", "step": 16300, "bytes_seen": 534118400, "loss": 2.5178885364532473, "throughput_bytes_per_second": 110924.08606043973, "gpu_hours": 1.3375473938219442, "lr": 0.0003076739338408002, "base_loss": 1.4180995225906372, "on_policy_rollout_loss": 3.1411287784576416, "rollout_steps": 2}
|
| 164 |
+
{"model": "R1", "step": 16400, "bytes_seen": 537395200, "loss": 2.438144643306732, "throughput_bytes_per_second": 110938.19867180336, "gpu_hours": 1.34558201081, "lr": 0.00030485914703893603, "base_loss": 0.9823269844055176, "on_policy_rollout_loss": 3.1857120990753174, "rollout_steps": 2}
|
| 165 |
+
{"model": "R1", "step": 16500, "bytes_seen": 540672000, "loss": 2.507713339328766, "throughput_bytes_per_second": 110952.23477643245, "gpu_hours": 1.3536155172475, "lr": 0.0003020471139095426, "base_loss": 1.0064581632614136, "on_policy_rollout_loss": 3.1619043350219727, "rollout_steps": 2}
|
| 166 |
+
{"model": "R1", "step": 16600, "bytes_seen": 543948800, "loss": 2.445737135410309, "throughput_bytes_per_second": 110966.20434856342, "gpu_hours": 1.36164780778, "lr": 0.00029923814245403225, "base_loss": 1.1980760097503662, "on_policy_rollout_loss": 3.1592049598693848, "rollout_steps": 2}
|
| 167 |
+
{"model": "R1", "step": 16700, "bytes_seen": 547225600, "loss": 2.437861204147339, "throughput_bytes_per_second": 110980.02453651758, "gpu_hours": 1.3696799198408334, "lr": 0.0002964325403384724, "base_loss": 1.1587449312210083, "on_policy_rollout_loss": 3.094273090362549, "rollout_steps": 2}
|
| 168 |
+
{"model": "R1", "step": 16800, "bytes_seen": 550502400, "loss": 2.450496289730072, "throughput_bytes_per_second": 110993.76752021418, "gpu_hours": 1.3777109899930555, "lr": 0.0002936306148598877, "base_loss": 1.1685312986373901, "on_policy_rollout_loss": 3.187481164932251, "rollout_steps": 2}
|
| 169 |
+
{"model": "R1", "step": 16900, "bytes_seen": 553779200, "loss": 2.4372637462615967, "throughput_bytes_per_second": 111007.61858147415, "gpu_hours": 1.3857387224522222, "lr": 0.00029083267291260144, "base_loss": 1.1884676218032837, "on_policy_rollout_loss": 3.096510887145996, "rollout_steps": 2}
|
| 170 |
+
{"model": "R1", "step": 17000, "bytes_seen": 557056000, "loss": 2.4655599713325502, "throughput_bytes_per_second": 111021.11371277596, "gpu_hours": 1.3937689201902776, "lr": 0.00028803902095462136, "base_loss": 1.0301244258880615, "on_policy_rollout_loss": 3.075671434402466, "rollout_steps": 2}
|