super_chat_20260306220449_6a67stalesuper_chat3.36M params6m 15s elapsed · Updated 38d ago
4L / 192D / 6H · helios · bpe-chat-4k · adamw· Created Mar 6, 2026 10:06 PM
Step 550 / 50,0001.1%
5.4630
Loss?
5.3799
Best Loss?
-34.5% from start
6.7934
Val Loss?
best: 6.7934
1.00e-3
Learning Rate?
32,781
Throughput?
tok/s (avg)
125
Speed?
ms/iter (avg)
1.530
Grad Norm?
avg: 2.358
2.25M
Tokens
processed
Loss Curve ? click any chart to add markers
?
?
?
?
Architecture
Layers?4
Embedding?192
Heads?6
Vocab?4,000
Context?256
Dropout?0
Parameters?3.36M
Training Config
Total iters?50,000
Batch size?8
Max LR?0.001
Optimizer?adamw
Backend?helios
Tokenizer?bpe-chat-4k
Seed?42
Weight decay?0.1
Grad clip?1
Eval interval?500
Throughput (tok/s)
Step Time (ms/iter)
GPU & VRAM
Perplexity
Train/Val Gap
No validation data
Learning Rate
Grad Norm
Smoothed Loss (EMA)
Loss Velocity
Gradient Clipping
GPU Operations
Step Time Breakdown
No timing data
Timing Phase Lines
No timing data
Backward / Forward Ratio
No timing data
Checkpoints (0) ?
No checkpoints saved
Sample Generations (1)
#CheckpointPrompt (preview)Generated
1-<|user|> Hello, how are you? <|assistant|>38d ago
Prompt
<|user|> Hello, how are you? <|assistant|>
Output
<|user|> Hello, how are you? <|assistant|>of eninit a an inand in the is inand inin in
Model Config (JSON)
{
"vocabSize": 4000,
"blockSize": 256,
"nLayer": 4,
"nEmbd": 192,
"nHead": 6,
"dropout": 0,
"ffnActivation": "gelu"
}Training Config (JSON)
{
"iters": 50000,
"batchSize": 8,
"lr": 0.001,
"lrMin": 0.0001,
"warmupIters": 500,
"beta1": 0.9,
"beta2": 0.95,
"eps": 1e-8,
"weightDecay": 0.1,
"gradClip": 1,
"evalInterval": 500,
"evalIters": 10,
"seed": 42,
"backend": "helios",
"tokenizer": "bpe-chat-4k",
"optimizer": "adamw",
"logLevel": "info",
"logEvery": 1,
"trace": false,
"gradAccumSteps": 2,
"sampleInterval": 500,
"spikeThreshold": 0,
"syncEvery": 0,
"gcEvery": 0,
"packed": true,
"symbio": false,
"symbioConfig": null
}