kacperwikiel commited on
Commit
c7eb0e9
·
verified ·
1 Parent(s): 801e32b

Update tokenizer, configuration and training reports

Browse files
reports/run/metrics.jsonl CHANGED
@@ -1198,3 +1198,7 @@
1198
  {"step": 23960, "tokens": 12561940480, "loss": 2.7447614669799805, "tokens_per_second": 997427.3005360373, "lr": 0.0015, "gradient_norm": 0.15338551998138428, "benchmark": false}
1199
  {"step": 23980, "tokens": 12572426240, "loss": 2.7036411285400392, "tokens_per_second": 741396.487194083, "lr": 0.0015, "gradient_norm": 0.1688719540834427, "benchmark": false}
1200
  {"step": 24000, "tokens": 12582912000, "loss": 2.718919372558594, "tokens_per_second": 982995.1583360904, "lr": 0.0015, "gradient_norm": 0.16046124696731567, "benchmark": false}
 
 
 
 
 
1198
  {"step": 23960, "tokens": 12561940480, "loss": 2.7447614669799805, "tokens_per_second": 997427.3005360373, "lr": 0.0015, "gradient_norm": 0.15338551998138428, "benchmark": false}
1199
  {"step": 23980, "tokens": 12572426240, "loss": 2.7036411285400392, "tokens_per_second": 741396.487194083, "lr": 0.0015, "gradient_norm": 0.1688719540834427, "benchmark": false}
1200
  {"step": 24000, "tokens": 12582912000, "loss": 2.718919372558594, "tokens_per_second": 982995.1583360904, "lr": 0.0015, "gradient_norm": 0.16046124696731567, "benchmark": false}
1201
+ {"step": 24020, "tokens": 12593397760, "loss": 2.7086679458618166, "tokens_per_second": 822883.7265012966, "lr": 0.0015, "gradient_norm": 0.1530439257621765, "benchmark": false}
1202
+ {"step": 24040, "tokens": 12603883520, "loss": 2.7279077529907227, "tokens_per_second": 980074.5210769067, "lr": 0.0015, "gradient_norm": 0.1471797525882721, "benchmark": false}
1203
+ {"step": 24060, "tokens": 12614369280, "loss": 2.7342641830444334, "tokens_per_second": 987047.5209221236, "lr": 0.0015, "gradient_norm": 0.15901459753513336, "benchmark": false}
1204
+ {"step": 24080, "tokens": 12624855040, "loss": 2.715492057800293, "tokens_per_second": 975792.0311073642, "lr": 0.0015, "gradient_norm": 0.15931537747383118, "benchmark": false}
reports/run/validation.jsonl CHANGED
@@ -21,3 +21,4 @@
21
  {"step": 21000, "source_validation_loss": {"fineweb_edu": 2.9123802185058594, "dclm_edu": 2.9512526988983154, "cosmopedia": 1.9636504650115967, "finemath": 2.145289421081543}}
22
  {"step": 22000, "source_validation_loss": {"fineweb_edu": 2.906125783920288, "dclm_edu": 2.946148633956909, "cosmopedia": 1.962751865386963, "finemath": 2.1453704833984375}}
23
  {"step": 23000, "source_validation_loss": {"fineweb_edu": 2.9029648303985596, "dclm_edu": 2.9421579837799072, "cosmopedia": 1.955125093460083, "finemath": 2.1363375186920166}}
 
 
21
  {"step": 21000, "source_validation_loss": {"fineweb_edu": 2.9123802185058594, "dclm_edu": 2.9512526988983154, "cosmopedia": 1.9636504650115967, "finemath": 2.145289421081543}}
22
  {"step": 22000, "source_validation_loss": {"fineweb_edu": 2.906125783920288, "dclm_edu": 2.946148633956909, "cosmopedia": 1.962751865386963, "finemath": 2.1453704833984375}}
23
  {"step": 23000, "source_validation_loss": {"fineweb_edu": 2.9029648303985596, "dclm_edu": 2.9421579837799072, "cosmopedia": 1.955125093460083, "finemath": 2.1363375186920166}}
24
+ {"step": 24000, "source_validation_loss": {"fineweb_edu": 2.9004201889038086, "dclm_edu": 2.94045090675354, "cosmopedia": 1.9505376815795898, "finemath": 2.12982439994812}}