A
Alpha
20260223223252_r4rgcompletedchat6.84M params1h 29m elapsed · Updated 48d ago
6L / 256D / 8H · helios · bpe-4k · adamw· Created Feb 23, 2026 10:33 PM
Step 17,500 / 50,00035.0%
4.3293
Loss?
3.9325
Best Loss?
-48.1% from start
4.7479
Val Loss?
best: 4.7188
2.78e-5
Learning Rate?
7,751
Throughput?
tok/s (avg)
529
Speed?
ms/iter (avg)
1209.090
Grad Norm?
avg: 1250.120
40.96M
Tokens
processed
63ms
Forward
12% of step
418ms
Backward
79% of step
32ms
GPU Sync
6% of step
949
GPU Ops
per step
1.0%
MFU
model FLOPS util
6.7x
Bwd/Fwd
ratio
Loss Curve ? click any chart to add markers
?
?
?
?
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?0.00003
Optimizer?adamw
Backend?helios
Tokenizer?bpe-4k
Seed?42
Weight decay?0.1
Grad clip?5
Eval interval?2500
Throughput (tok/s)
Step Time (ms/iter)
GPU & VRAM
Perplexity
Train/Val Gap
Learning Rate
Grad Norm
Smoothed Loss (EMA)
Loss Velocity
Gradient Clipping
No clipping data
GPU Operations
Step Time Breakdown
Forward
Backward
Grad Norm
Optimizer
GPU Sync
Data
Timing Phase Lines
Backward / Forward Ratio
Checkpoints (0) ?
No checkpoints saved
Sample Generations (3)
#CheckpointPrompt (preview)Generated
1-<|user|> Hello, how are you? <|assistant|>49d ago
Prompt
<|user|> Hello, how are you? <|assistant|>
Output
<|user|> Hello, how are you? <|assistant|> <|user|> AT <|assistant|> T <|user|> woh is oye, 10 <|user|> SERANES <|user|> what you <|assistant|> yyoh <|assistant|> I da go <|assistant|> hi <|assistant|> k<|end_
2-<|user|> What do you like to do for fun? <|assistant|>49d ago
Prompt
<|user|> What do you like to do for fun? <|assistant|>
Output
<|user|> What do you like to do for fun? <|assistant|> <|assistant|> I’re for i use i <|assistant|> i didn't think its not it just my 20+ <|assistant|> wing it and not your stened renglie 133 <|user|> ITomnuh i
3-<|user|> Tell me about yourself. <|assistant|>49d ago
Prompt
<|user|> Tell me about yourself. <|assistant|>
Output
<|user|> Tell me about yourself. <|assistant|> <|assistant|> I can a da <|assistant|> What the da wined is one i would got the def<|end_of_text|> <|user|> I just the servers a feren <|user|> i feel 1<|en
Model Config (JSON)
{
  "vocabSize": 4000,
  "blockSize": 256,
  "nLayer": 6,
  "nEmbd": 256,
  "nHead": 8,
  "dropout": 0
}
Training Config (JSON)
{
  "iters": 50000,
  "batchSize": 16,
  "lr": 0.00003,
  "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": true,
  "gradAccumSteps": 1,
  "sampleInterval": 2500,
  "spikeThreshold": 500
}