A
Alpha
20260219032338_r8ypcompletednovels726.3K params9m 52s elapsed · Updated 6d ago
3L / 96D / 4H · cpu_ref · bpe · adamw· Created Feb 19, 2026 3:24 AM
Step 99 / 10099.0%
6.9420
Loss?
6.8673
Best Loss?
-9.0% from start
6.9593
Val Loss?
best: 6.9593
1.83e-7
Learning Rate?
96
Throughput?
tok/s (avg)
6,127
Speed?
ms/iter (avg)
0.492
Grad Norm?
avg: 0.500
55.9K
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?3
Embedding?96
Heads?4
Vocab?2,000
Context?96
Dropout?0
Parameters?726.3K
Training Config
Total iters?100
Batch size?6
Max LR?0.0006
Optimizer?adamw
Backend?cpu_ref
Tokenizer?bpe
Seed?42
Weight decay?0.01
Grad clip?1
Eval interval?50
GPU & VRAM
No GPU data
Learning Rate
Grad Norm
Step Time Breakdown
Step Time Breakdown
No timing data
Checkpoints (0) ?
No checkpoints saved
{
  "vocabSize": 2000,
  "blockSize": 96,
  "nLayer": 3,
  "nEmbd": 96,
  "nHead": 4,
  "dropout": 0
}
{
  "iters": 100,
  "batchSize": 6,
  "lr": 0.0006,
  "beta1": 0.9,
  "beta2": 0.999,
  "eps": 1e-8,
  "weightDecay": 0.01,
  "gradClip": 1,
  "evalInterval": 50,
  "evalIters": 10,
  "seed": 42,
  "backend": "cpu_ref",
  "tokenizer": "bpe",
  "optimizer": "adamw",
  "logLevel": "info",
  "trace": false
}