A
Alpha
20260224020120_8kr9completedchat6.84M params1h 28m elapsed · Updated 1d ago
6L / 256D / 8H · helios · bpe-4k · adamw· Created Feb 24, 2026 2:02 AM
Step 11,005 / 50,00022.0%
6.6415
Loss?
5.2129
Best Loss?
-20.3% from start
6.5857
Val Loss?
best: 6.2235
2.74e-4
Learning Rate?
7,731
Throughput?
tok/s (avg)
530
Speed?
ms/iter (avg)
1382600.450
Grad Norm?
avg: 1402494.703
40.96M
Tokens
processed
63ms
Forward
12% of step
407ms
Backward
77% of step
25ms
GPU Sync
5% of step
1,064
GPU Ops
per step
1.0%
MFU
model FLOPS util
6.5x
Bwd/Fwd
ratio
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?50,000
Batch size?16
Max LR?0.0003
Optimizer?adamw
Backend?helios
Tokenizer?bpe-4k
Seed?42
Weight decay?0.1
Grad clip?5
Eval interval?2500
GPU & VRAM
Learning Rate
Grad Norm
Step Time Breakdown
Step Time Breakdown
Checkpoints (0) ?
No checkpoints saved
Sample Generations (3)
#CheckpointPrompt (preview)Generated
1-<|user|> Hello, how are you? <|assistant|>1d ago
Prompt
<|user|> Hello, how are you? <|assistant|>
Output
<|user|> Hello, how are you? <|assistant|>a TR<|assistant|> M it my the <|assistant|> agtdan <|assistant|> GOis S <|user|> Sopi<|end_of_text|> <|user|> LObut the Ies <|assistant|> my the <|assistant|> <|assistant|> PNOA <|assistant|>
2-<|user|> What do you like to do for fun? <|assistant|>1d ago
Prompt
<|user|> What do you like to do for fun? <|assistant|>
Output
<|user|> What do you like to do for fun? <|assistant|>etari <|assistant|> i what i it <|assistant|> ’wata that <|assistant|> S <|assistant|> <|assistant|> OWRit what bt <|user|> <|assistant|> OU<|assistant|> L what ast I EY E i<|end_of_text|> <|user|>
3-<|user|> Tell me about yourself. <|assistant|>1d ago
Prompt
<|user|> Tell me about yourself. <|assistant|>
Output
<|user|> Tell me about yourself. <|assistant|> <|assistant|> Mwm <|assistant|> <|user|> <|assistant|> I your I <|assistant|> RI i <|assistant|> I Euleo is I a<|end_of_text|> <|user|> <|assistant|> Yaa <|user|> NE<|user|> CNT E<|user|> R<|user|> <|assistant|> <|assistant|> 
{
  "vocabSize": 4000,
  "blockSize": 256,
  "nLayer": 6,
  "nEmbd": 256,
  "nHead": 8,
  "dropout": 0
}
{
  "iters": 50000,
  "batchSize": 16,
  "lr": 0.0003,
  "lrMin": 0.000005,
  "warmupIters": 500,
  "beta1": 0.9,
  "beta2": 0.95,
  "eps": 0.000001,
  "weightDecay": 0.1,
  "gradClip": 5,
  "evalInterval": 2500,
  "evalIters": 10,
  "seed": 42,
  "backend": "helios",
  "tokenizer": "bpe-4k",
  "optimizer": "adamw",
  "logLevel": "info",
  "trace": false,
  "gradAccumSteps": 1,
  "sampleInterval": 2500,
  "spikeThreshold": 0
}