step 3600 | val_ppl=12.20

README.md

@@ -25,7 +25,7 @@ Small language model (3.0M parameters) trained from scratch.
 | Total parameters | 3.000M |
 
 ## Training
-- Tokens seen: 3,352,166,400
+- Tokens seen: 3,771,187,200
 - Val loss: 2.5017
 - Val PPL: 12.20
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b920a59624b0d252475ee89f2eb6d0d97020d5e23312607c88801345302de0a0
+oid sha256:500a97da454521a720ec0095d6dff92658424b984d178ae1502e7cfc85364b34
 size 12006544

resume/ckpt.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a43f75896e2dc9137d50dc206ae6f3e8c7f86fedbe87257180cc7580f914a826
+oid sha256:bff85951328748b62934217013ec7499b1e7bf4178dcb10a6be5fdcdcff3f3eb
 size 36728047

resume/latest_step.txt

@@ -1 +1 @@
-3200
+3600

training_meta.json

@@ -1,8 +1,8 @@
 {
-  "step": 3200,
+  "step": 3600,
   "val_loss": 2.501656575469971,
   "val_ppl": 12.202691906453575,
   "params_M": 3.0,
-  "pushed_at": "2026-07-01T15:58:56.609834",
-  "tokens_seen": 3352166400
+  "pushed_at": "2026-07-01T16:49:32.162250",
+  "tokens_seen": 3771187200
 }