File size: 1,294 Bytes
a2966b3 | 1 2 3 4 5 6 7 8 9 10 |
Loading checkpoint shards: 0%| | 0/3 [00:00<?, ?it/s]
Loading checkpoint shards: 33%|ββββ | 1/3 [00:00<00:00, 9.25it/s]
Loading checkpoint shards: 67%|βββββββ | 2/3 [00:00<00:00, 8.94it/s]
Loading checkpoint shards: 100%|ββββββββββ| 3/3 [00:00<00:00, 9.74it/s]
baseline c4=4.3505
Loading checkpoint shards: 0%| | 0/3 [00:00<?, ?it/s]
Loading checkpoint shards: 33%|ββββ | 1/3 [00:00<00:00, 9.18it/s]
Loading checkpoint shards: 67%|βββββββ | 2/3 [00:00<00:00, 8.99it/s]
Loading checkpoint shards: 100%|ββββββββββ| 3/3 [00:00<00:00, 9.86it/s]
uniform_keep25 c4=5.0687 [dead 0/1024, kept 262144, params 2.09B]
Loading checkpoint shards: 0%| | 0/3 [00:00<?, ?it/s]
Loading checkpoint shards: 33%|ββββ | 1/3 [00:00<00:00, 9.10it/s]
Loading checkpoint shards: 67%|βββββββ | 2/3 [00:00<00:00, 8.79it/s]
Loading checkpoint shards: 100%|ββββββββββ| 3/3 [00:00<00:00, 9.60it/s]
uniform_keep75 c4=4.6639 [dead 0/1024, kept 786432, params 5.31B]
decision gate (variant NLL - glean NLL; positive = glean better):
saved -> outputs/uniform_control_materialize/results.json
|