step 5600 | val_ppl=11.05

README.md

@@ -25,9 +25,9 @@ Small language model (9.9M parameters) trained from scratch.
 | Total parameters | 9.853M |
 
 ## Training
-- Tokens seen: 5,447,270,400
-- Val loss: 2.4072
-- Val PPL: 11.10
+- Tokens seen: 5,866,291,200
+- Val loss: 2.4024
+- Val PPL: 11.05
 
 ## Usage
 ```python

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20ff95719d6121558575337196a10ba0f928e986f999ac4f80aa3e8855c05a6b
+oid sha256:d72c1f17e08af6e93e10d46c14123b1b753c022f4222f492c1f6d09a7170b896
 size 39421616

resume/ckpt.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:226ab947999a0de9b12295aa18e40b1952ecdaffef7a0dee8f40f20498d13b3a
+oid sha256:8277ce113e0fcbf62e0b9e360c65b2e4b595cecd4d56da8fd42f1d34746e7a68
 size 119812327

resume/latest_step.txt

@@ -1 +1 @@
-5200
+5600

training_meta.json

@@ -1,8 +1,8 @@
 {
-  "step": 5200,
-  "val_loss": 2.4072011318206785,
-  "val_ppl": 11.102842224699165,
+  "step": 5600,
+  "val_loss": 2.402378048706055,
+  "val_ppl": 11.04942122436039,
   "params_M": 9.853,
-  "pushed_at": "2026-06-18T09:48:36.577662",
-  "tokens_seen": 5447270400
+  "pushed_at": "2026-06-18T11:49:11.294351",
+  "tokens_seen": 5866291200
 }