novels-5hrcompletednovels2.50M params5h 30m elapsed · Updated 6m ago
6L / 128D / 4H · cpu_ref · bpe · adamw· Created Feb 19, 2026 12:24 AM
Step 900 / 900100.0%
5.3346
Loss?
5.2128
Best Loss?
-30.0% from start
5.5975
Val Loss?
best: 5.5691
1.13e-9
Learning Rate?
52
Throughput?
tok/s (avg)
20,227
Speed?
ms/iter (avg)
1.522
Grad Norm?
avg: 1.775
921.6K
Tokens
processed
Loss Curve ?
Validation / selection
Run diagnostics
Gradient + Loss
Train/val loss with validation, warmup, overfit, and instability markers.
Search-aware view
Architecture
Layers?6
Embedding?128
Heads?4
Vocab?2,000
Context?128
Dropout?0
Parameters?2.50M
Training Config
Total iters?900
Batch size?8
Max LR?0.0003
Optimizer?adamw
Backend?cpu_ref
Tokenizer?bpe
Seed?42
Weight decay?0.01
Grad clip?1
Eval interval?100
GPU & VRAM
No GPU data
Learning Rate
Grad Norm
Step Time Breakdown
Step Time Breakdown
No timing data
Checkpoints (9) ?
StepFilenameSizeCreated
{
"vocabSize": 2000,
"blockSize": 128,
"nLayer": 6,
"nEmbd": 128,
"nHead": 4,
"dropout": 0
}{
"iters": 900,
"batchSize": 8,
"lr": 0.0003,
"beta1": 0.9,
"beta2": 0.999,
"eps": 1e-8,
"weightDecay": 0.01,
"gradClip": 1,
"evalInterval": 100,
"evalIters": 10,
"seed": 42,
"backend": "cpu_ref",
"tokenizer": "bpe",
"optimizer": "adamw",
"logLevel": "info",
"trace": false
}