Joint training checkpoint step 3175

README.md

@@ -10,7 +10,7 @@ tags: [tabular-data, masked-lm, token-classification, experimental]
 Experimental budget-limited mmBERT-base joint fine-tuning checkpoint.
 
 One shared encoder, 18 category BIO outputs, insertion outputs, and the pretrained MLM vocabulary head.
-Completed optimizer updates: **1210**. This checkpoint is a research baseline, not a validated automatic cleaning system.
+Completed optimizer updates: **3175**. This checkpoint is a research baseline, not a validated automatic cleaning system.
 
 ## Use and resume
 
@@ -40,10 +40,10 @@ TabFix fine-tuning contributions are CC BY-NC 4.0; commercial use requires the o
   "start_step": 1200,
   "device": "cuda",
   "parameters": 307843915,
-  "step": 1210,
-  "last_detection_loss": 0.0012828651815652847,
-  "last_correction_loss": 1.3493580809154082e-05,
-  "training_elapsed_seconds": 4.6002243840000006,
+  "step": 3175,
+  "last_detection_loss": 0.002469167113304138,
+  "last_correction_loss": 6.357827686542805e-08,
+  "training_elapsed_seconds": 622.852989595,
   "exclusions": {}
 }
 ```

checkpoint.json

@@ -1,19 +1,19 @@
 {
   "version": 1,
-  "step": 1210,
+  "step": 3175,
   "dataset_revision": "7da5aa80bc2a68a375930f3bd3ed2f7fdda0152f",
   "base_revision": "c5955035435e2bf121cde7f3c8863ef52ff35d82",
   "files": {
-    "model.safetensors": "501011bd841a613bc383abc44d57860fb5a9ab5a9c758799a9ea20aa22ed8bbe",
+    "model.safetensors": "ccfb450d26fcb498772003366ee7b8da5e0ab6293df6c48fedc51c90316e9b76",
     "config.json": "9e105ce4ed25ed2d6be2443eeb69932a530668a248ce3211d43a89a136d42531",
     "tokenizer_config.json": "84c64b446e2719fddccf2f753e538a33e5c35bc616c80b47e2622c6e0debafeb",
     "tokenizer.json": "47e408472c05385fea5fb9f8b02bbd838d55da8f15a22c863a99fd31d89243f5",
-    "training_state.pt": "cf6e7eeadbbc1493b05373dcf0d8b8192f4f24b9bcb6f6a2a222065033378e49",
+    "training_state.pt": "dbb5c8d9d66f72a3625428b8b042fb05c05f808f3c7d7a4f1fa42430bbc62903",
     "run_config.json": "c968dd374ed2ec1d246dc7a47e7625a430b673281265cebc483991736121aaa4",
-    "metrics.json": "de7326eedeb99949b41b41a9c0aa61149e74c2f71d2e9e78d5b7b6525eaa8358",
+    "metrics.json": "bec1ecb9c02237e5f9a44bfb19442a5a1c77187a37fd99bc5af2ea51236dfab1",
     "categories.json": "d95af1bcb70bc78d53297d4e57909776773fff76329098ece751e0734f03b29b",
     "train_tabfix.py": "1f6d80de49a3bb5d89affe2c82b7e558f9d983322e6f5a2747aae58ecef42264",
-    "README.md": "23057992614ba35fd9f8e1cb3bf91618d62d282443a878ab87b0ad11ac431584",
+    "README.md": "f0d6defe252d59222067cb2a0213a2b089e84ede46ae7c60506aebfe12e85257",
     "LICENSE": "b6c1c7bb5aae03bb6b3396fc7c65cbe7306c422ed424b0b7b687d9a50cc15412",
     "BASE_MODEL_CARD.md": "35724415037ee3159aa733c16050620cc19a76219112a6ceeb3949572e0b37e9"
   },

metrics.json

@@ -7,9 +7,9 @@
   "start_step": 1200,
   "device": "cuda",
   "parameters": 307843915,
-  "step": 1210,
-  "last_detection_loss": 0.0012828651815652847,
-  "last_correction_loss": 1.3493580809154082e-05,
-  "training_elapsed_seconds": 4.6002243840000006,
+  "step": 3175,
+  "last_detection_loss": 0.002469167113304138,
+  "last_correction_loss": 6.357827686542805e-08,
+  "training_elapsed_seconds": 622.852989595,
   "exclusions": {}
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:501011bd841a613bc383abc44d57860fb5a9ab5a9c758799a9ea20aa22ed8bbe
+oid sha256:ccfb450d26fcb498772003366ee7b8da5e0ab6293df6c48fedc51c90316e9b76
 size 1231390724

training_state.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf6e7eeadbbc1493b05373dcf0d8b8192f4f24b9bcb6f6a2a222065033378e49
-size 2462881846
+oid sha256:dbb5c8d9d66f72a3625428b8b042fb05c05f808f3c7d7a4f1fa42430bbc62903
+size 2462881910