step 5200 | val_ppl=11.10

README.md

@@ -25,9 +25,9 @@ Small language model (9.9M parameters) trained from scratch.
 | Total parameters | 9.853M |
 
 ## Training
-- Tokens seen: 5,028,249,600
-- Val loss: 2.4145
-- Val PPL: 11.18
+- Tokens seen: 5,447,270,400
+- Val loss: 2.4072
+- Val PPL: 11.10
 
 ## Usage
 ```python

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e34e57b55af297cecb8a522368b79288dd25ac39a7d39b1336f31dc51769fd9
+oid sha256:20ff95719d6121558575337196a10ba0f928e986f999ac4f80aa3e8855c05a6b
 size 39421616

resume/ckpt.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e542a611020536fbbda2187c18314a69cae02cbb51fce840799a4e4e35c923e0
+oid sha256:226ab947999a0de9b12295aa18e40b1952ecdaffef7a0dee8f40f20498d13b3a
 size 119812327

resume/latest_step.txt

@@ -1 +1 @@
-4800
+5200

training_meta.json

@@ -1,8 +1,8 @@
 {
-  "step": 4800,
-  "val_loss": 2.4144561973571776,
-  "val_ppl": 11.18368698508662,
+  "step": 5200,
+  "val_loss": 2.4072011318206785,
+  "val_ppl": 11.102842224699165,
   "params_M": 9.853,
-  "pushed_at": "2026-06-18T07:47:49.668654",
-  "tokens_seen": 5028249600
+  "pushed_at": "2026-06-18T09:48:36.577662",
+  "tokens_seen": 5447270400
 }