step 13200 | val_ppl=9.47

README.md

@@ -25,7 +25,7 @@ Small language model (3.0M parameters) trained from scratch.
 | Total parameters | 3.000M |
 
 ## Training
-- Tokens seen: 13,408,665,600
+- Tokens seen: 13,827,686,400
 - Val loss: 2.2479
 - Val PPL: 9.47
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d47eaa983c7793bd06901c5a1baaa0405bb4d21718b33feacb0dcbd7ec8c750
+oid sha256:e2bbf1a453ed14c2071e0cb91a3fd41111b926d2381af584fbbcd8cd93be4213
 size 12006544

resume/ckpt.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8a204da107184eae47c99925beefbd7709c6643c3e51b0f5e4dbbb88c3fa168
+oid sha256:2f70b51ef306b27c7e56d9d75588d9a80bc4a56c3ef015cae212dcab8b26332f
 size 36728175

resume/latest_step.txt

@@ -1 +1 @@
-12800
+13200

training_meta.json

@@ -1,8 +1,8 @@
 {
-  "step": 12800,
+  "step": 13200,
   "val_loss": 2.2479474616241455,
   "val_ppl": 9.468281866294314,
   "params_M": 3.0,
-  "pushed_at": "2026-07-03T19:21:06.751410",
-  "tokens_seen": 13408665600
+  "pushed_at": "2026-07-03T20:08:24.851598",
+  "tokens_seen": 13827686400
 }