Backup runs/R1_aligned_v2/training.jsonl (38,144 bytes)
Browse files
byte_repair_v1/runs/R1_aligned_v2/training.jsonl
CHANGED
|
@@ -118,3 +118,13 @@
|
|
| 118 |
{"model": "R1", "step": 11800, "bytes_seen": 386662400, "loss": 2.4708056378364565, "throughput_bytes_per_second": 110983.106590637, "gpu_hours": 0.9677709114630555, "lr": 0.0004321943696898156, "base_loss": 0.9544661641120911, "on_policy_rollout_loss": 3.1608774662017822, "rollout_steps": 2}
|
| 119 |
{"model": "R1", "step": 11900, "bytes_seen": 389939200, "loss": 2.4611511039733887, "throughput_bytes_per_second": 111001.81655335437, "gpu_hours": 0.9758078543911111, "lr": 0.0004295733113838352, "base_loss": 1.462003469467163, "on_policy_rollout_loss": 3.096226692199707, "rollout_steps": 2}
|
| 120 |
{"model": "R1", "step": 12000, "bytes_seen": 393216000, "loss": 2.4492714405059814, "throughput_bytes_per_second": 111020.46700806051, "gpu_hours": 0.9838426157830555, "lr": 0.0004269413468334722, "base_loss": 1.4526588916778564, "on_policy_rollout_loss": 3.131598949432373, "rollout_steps": 2}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
{"model": "R1", "step": 11800, "bytes_seen": 386662400, "loss": 2.4708056378364565, "throughput_bytes_per_second": 110983.106590637, "gpu_hours": 0.9677709114630555, "lr": 0.0004321943696898156, "base_loss": 0.9544661641120911, "on_policy_rollout_loss": 3.1608774662017822, "rollout_steps": 2}
|
| 119 |
{"model": "R1", "step": 11900, "bytes_seen": 389939200, "loss": 2.4611511039733887, "throughput_bytes_per_second": 111001.81655335437, "gpu_hours": 0.9758078543911111, "lr": 0.0004295733113838352, "base_loss": 1.462003469467163, "on_policy_rollout_loss": 3.096226692199707, "rollout_steps": 2}
|
| 120 |
{"model": "R1", "step": 12000, "bytes_seen": 393216000, "loss": 2.4492714405059814, "throughput_bytes_per_second": 111020.46700806051, "gpu_hours": 0.9838426157830555, "lr": 0.0004269413468334722, "base_loss": 1.4526588916778564, "on_policy_rollout_loss": 3.131598949432373, "rollout_steps": 2}
|
| 121 |
+
{"model": "R1", "step": 12100, "bytes_seen": 396492800, "loss": 2.4502150559425355, "throughput_bytes_per_second": 110815.19038613913, "gpu_hours": 0.9938789845066667, "lr": 0.0004242987643172642, "base_loss": 1.494910478591919, "on_policy_rollout_loss": 3.2663891315460205, "rollout_steps": 2}
|
| 122 |
+
{"model": "R1", "step": 12200, "bytes_seen": 399769600, "loss": 2.4532784962654115, "throughput_bytes_per_second": 110835.09752073711, "gpu_hours": 1.0019128741266667, "lr": 0.0004216458532767325, "base_loss": 1.5393168926239014, "on_policy_rollout_loss": 3.14866042137146, "rollout_steps": 2}
|
| 123 |
+
{"model": "R1", "step": 12300, "bytes_seen": 403046400, "loss": 2.463639452457428, "throughput_bytes_per_second": 110854.82418842864, "gpu_hours": 1.0099455224702776, "lr": 0.0004189829042846795, "base_loss": 1.2338110208511353, "on_policy_rollout_loss": 3.1229124069213867, "rollout_steps": 2}
|
| 124 |
+
{"model": "R1", "step": 12400, "bytes_seen": 406323200, "loss": 2.457118091583252, "throughput_bytes_per_second": 110874.15257671375, "gpu_hours": 1.017978969241389, "lr": 0.0004163102090133623, "base_loss": 1.4338624477386475, "on_policy_rollout_loss": 3.091977596282959, "rollout_steps": 2}
|
| 125 |
+
{"model": "R1", "step": 12500, "bytes_seen": 409600000, "loss": 2.465863835811615, "throughput_bytes_per_second": 110893.434173783, "gpu_hours": 1.0260100485252777, "lr": 0.0004136280602025461, "base_loss": 0.8762134313583374, "on_policy_rollout_loss": 3.1383793354034424, "rollout_steps": 2}
|
| 126 |
+
{"model": "R1", "step": 12600, "bytes_seen": 412876800, "loss": 2.44458687543869, "throughput_bytes_per_second": 110912.12538822734, "gpu_hours": 1.0340438396483334, "lr": 0.00041093675162744045, "base_loss": 1.3126492500305176, "on_policy_rollout_loss": 3.11309814453125, "rollout_steps": 2}
|
| 127 |
+
{"model": "R1", "step": 12700, "bytes_seen": 416153600, "loss": 2.4554046893119814, "throughput_bytes_per_second": 110930.78011249776, "gpu_hours": 1.0420752662605555, "lr": 0.0004082365780665215, "base_loss": 1.1219037771224976, "on_policy_rollout_loss": 3.0891125202178955, "rollout_steps": 2}
|
| 128 |
+
{"model": "R1", "step": 12800, "bytes_seen": 419430400, "loss": 2.454271745681763, "throughput_bytes_per_second": 110948.73187230066, "gpu_hours": 1.0501106455055556, "lr": 0.00040552783526924617, "base_loss": 0.9728826284408569, "on_policy_rollout_loss": 3.1253325939178467, "rollout_steps": 2}
|
| 129 |
+
{"model": "R1", "step": 12900, "bytes_seen": 422707200, "loss": 2.4613593316078184, "throughput_bytes_per_second": 110967.05014095297, "gpu_hours": 1.058139929983889, "lr": 0.0004028108199236575, "base_loss": 1.402837872505188, "on_policy_rollout_loss": 3.149467945098877, "rollout_steps": 2}
|
| 130 |
+
{"model": "R1", "step": 13000, "bytes_seen": 425984000, "loss": 2.4397303581237795, "throughput_bytes_per_second": 110984.7179251646, "gpu_hours": 1.0661728128072223, "lr": 0.0004000858296238893, "base_loss": 1.2418383359909058, "on_policy_rollout_loss": 3.0759060382843018, "rollout_steps": 2}
|