A
Alpha
20260223014834_63p2completedchat6.84M params5m 58s elapsed · Updated 1d ago
6L / 256D / 8H · helios · bpe-4k · adamw· Created Feb 23, 2026 1:49 AM
Step 842 / 30,0002.8%
6.6721
Loss?
5.1080
Best Loss?
-19.6% from start
8.5426
Val Loss?
best: 5.7704
1.00e-4
Learning Rate?
7,936
Throughput?
tok/s (avg)
529
Speed?
ms/iter (avg)
4719.308
Grad Norm?
avg: 1444.051
2.91M
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?256
Heads?8
Vocab?4,000
Context?256
Dropout?0
Parameters?6.84M
Training Config
Total iters?30,000
Batch size?16
Max LR?0.0001
Optimizer?adamw
Backend?helios
Tokenizer?bpe-4k
Seed?42
Weight decay?0.1
Grad clip?1
Eval interval?100
GPU & VRAM
Learning Rate
Grad Norm
Step Time Breakdown
Step Time Breakdown
No timing data
Checkpoints (0) ?
No checkpoints saved
Sample Generations (3)
#CheckpointPrompt (preview)Generated
1-<|user|> Hello, how are you? <|assistant|>2d ago
Prompt
<|user|> Hello, how are you? <|assistant|>
Output
<|user|> Hello, how are you? <|assistant|>but Wcfrom rcea to a fis to esome br<|user|> itf ding itto the es es lio from do moe? a a agal<|assistant|> o ts some <|et tfito dra
2-<|user|> What do you like to do for fun? <|assistant|>2d ago
Prompt
<|user|> What do you like to do for fun? <|assistant|>
Output
<|user|> What do you like to do for fun? <|assistant|>fto fa <|user|> ding brjust <|user|> gjust but can not s ? es <|user|> eo but <|user|> fds taring ? Tacan you from ci <|user|> but time TT|>more some al
3-<|user|> Tell me about yourself. <|assistant|>2d ago
Prompt
<|user|> Tell me about yourself. <|assistant|>
Output
<|user|> Tell me about yourself. <|assistant|>ealtedto , to in a some is ing your is altanbrjust is anaan c Wto <|to br<|just , and the ars i just altfin <|user|> cd_t
{
  "vocabSize": 4000,
  "blockSize": 256,
  "nLayer": 6,
  "nEmbd": 256,
  "nHead": 8,
  "dropout": 0
}
{
  "iters": 30000,
  "batchSize": 16,
  "lr": 0.0001,
  "beta1": 0.9,
  "beta2": 0.95,
  "eps": 1e-8,
  "weightDecay": 0.1,
  "gradClip": 1,
  "evalInterval": 100,
  "evalIters": 10,
  "seed": 42,
  "backend": "helios",
  "tokenizer": "bpe-4k",
  "optimizer": "adamw",
  "logLevel": "info",
  "trace": false
}