20260222141034_mdbpcompletedconcordance115.55M params1m 27s elapsed · Updated 1d ago
12L / 768D / 12H · helios · bpe-64k · adamw· Created Feb 22, 2026 2:10 PM
Step 1,000 / 5,00020.0%
7.2724
Loss?
7.2724
Best Loss?
-26.5% from start
-
Val Loss?
3.68e-5
Learning Rate?
1,409
Throughput?
tok/s (avg)
5,826
Speed?
ms/iter (avg)
798.432
Grad Norm?
avg: 75.253
122.9K
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?12
Embedding?768
Heads?12
Vocab?19,777
Context?256
Dropout?0
Parameters?115.55M
Training Config
Total iters?5,000
Batch size?32
Max LR?0.00004
Optimizer?adamw
Backend?helios
Tokenizer?bpe-64k
Seed?42
Weight decay?0.01
Grad clip?1
Eval interval?500
GPU & VRAM
Learning Rate
Grad Norm
Step Time Breakdown
Step Time Breakdown
No timing data
Checkpoints (0) ?
No checkpoints saved
{
"vocabSize": 19777,
"blockSize": 256,
"nLayer": 12,
"nEmbd": 768,
"nHead": 12,
"dropout": 0
}{
"iters": 5000,
"batchSize": 32,
"lr": 0.00004,
"beta1": 0.9,
"beta2": 0.999,
"eps": 1e-8,
"weightDecay": 0.01,
"gradClip": 1,
"evalInterval": 500,
"evalIters": 10,
"seed": 42,
"backend": "helios",
"tokenizer": "bpe-64k",
"optimizer": "adamw",
"logLevel": "info",
"trace": false
}