main chck_100M

.complete.json

@@ -2,13 +2,13 @@
   "files": {
     "config.json": "1cda17ac7f888bdbecee636a6a9ff3f6486ce02eee0871c5738ffaf859949b6b",
     "factor_readout.safetensors": "db1b6b3819d70b856ff2df0e502072bfdd9f2b273749522afa21ef55ae0bd7b8",
-    "model.safetensors": "2a34fc63f2470916568127f21eab885ecea0443dac8f4d7d10d1df3e49d764ff",
+    "model.safetensors": "c1dcf8e7852613dec57631a9fe7bfd78882bf5225d34b1de5111d46c25ef5634",
     "special_tokens_map.json": "2be97b602cd0e4a6c2874cbf03c0d1025b3af5474666da931bc04f6c23a9a39d",
     "tokenizer.json": "2b8d1b3f51c0d8f276a64ea6a69efa50b4d9899780f1b282b4d0dcb7be5bbb0f",
     "tokenizer_config.json": "8f4a998564d87fc85b27652f0b295677a1e03744bf400bfacc342b700cc3653a",
     "tolm.py": "a9c9469b58e93b7f0228e0c00442735bd0304278e549af37fd3766a902bb1d13",
-    "training_manifest.json": "e58549050b490a2d77aad3243ad397841752e07317816e5c6c348e37a6631b54"
+    "training_manifest.json": "f9f66184e56b73ce9d44d1d821079dc35c14099999298e39790bee1db2ff2c44"
   },
   "schema_version": 1,
-  "words_seen": 88000000
+  "words_seen": 100000000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a34fc63f2470916568127f21eab885ecea0443dac8f4d7d10d1df3e49d764ff
+oid sha256:c1dcf8e7852613dec57631a9fe7bfd78882bf5225d34b1de5111d46c25ef5634
 size 133365608

training_manifest.json

@@ -40,7 +40,8 @@
     "auxiliary_linguistic_training_words": 2000000,
     "brown_training_words": 1000000,
     "competition_status": "strict_small_conservative_accounting",
-    "conservative_counted_words": 99900000,
+    "conservative_corpus_accounting_words": 10000000,
+    "conservative_counted_words": 100000000,
     "corpus_induced_priors": true,
     "corpus_limit_words": 10000000,
     "evaluation_holdout_words": 1000000,
@@ -49,17 +50,17 @@
     "generated_words": 0,
     "induction_corpus_disjoint_from_lm_and_holdout": true,
     "leaderboard_checkpoint_words": 88000000,
-    "model_exposure_words": 88000000,
+    "model_exposure_words": 100000000,
     "model_training_corpus_words": 8900000,
     "ppmi_induction_words": 1000000,
     "relational_corpus_fraction": 0.025,
     "same_run_checkpoint_distribution": false,
     "signals_derived_from_current_model_input": false,
-    "strict_small_status": "within_both_books_without_baseline_precedent",
+    "strict_small_status": "corpus_10m_model_views_100m",
     "syntax_training_words": 0,
     "teacher_queries": 0,
     "tokenizer_training_words": 8900000,
-    "total_exposure_words": 99900000,
+    "total_exposure_words": 100000000,
     "track": "strict-small",
     "unique_corpus_words": 10000000,
     "within_100m_conservative_budget": true,
@@ -215,7 +216,7 @@
     "ema_decay": 0.9998,
     "epsilon": 1e-08,
     "exact_word_checkpoints": true,
-    "exposure_words": 88000000,
+    "exposure_words": 100000000,
     "final_lr_ratio": 0.1,
     "frequency_aware_masking": {
       "enabled": false,
@@ -281,7 +282,9 @@
       60000000,
       70000000,
       80000000,
-      88000000
+      88000000,
+      90000000,
+      100000000
     ],
     "schedule_total_words": 100000000,
     "span_max_length": 3,
@@ -304,5 +307,5 @@
     "z_loss_weight": 0.0001
   },
   "variant": "factorized_dense",
-  "words_seen": 88000000
+  "words_seen": 100000000
 }