stale
super_chat_20260306143657_ns34
55.77M parameter nanochat model — bpe-chat-4k tokenizer, 16L/512D/8H
Overview
55.77M
Parameters
6.9962
Final Loss
-
Best Val Loss
1092.5
Perplexity
299,008
Tokens Processed
0.0
Tokens/Param
1,215 tok/s
Avg Throughput
4m 8s
Training Time
Training Progress146 / 200 steps (73.0%)
Loss reduced by 16.6% from initial 8.3936
Dataset & Training
Domainnanochat
Tokenizerbpe-chat-4k
Total Iterations200
Batch Size4
Context Length512 tokens
Tokens per Batch2,048
Dataset Passes~1
Effective Tokens299,008
Training Pipeline
Warmupsteps 1–2
Learning rate warmup — model weights adjusting to data distribution
Loss: 8.394 → 8.396Linear LR warmup, gradient clipping
Rapid Descentsteps 2–61
Steepest loss reduction — model learning primary patterns
Loss: 8.396 → 6.767Cosine LR schedule, AdamW optimization
Refinementsteps 61–141
Diminishing returns — model fine-tuning subtler patterns
Loss: 6.767 → 7.082Lower LR, gradient accumulation
Convergencesteps 141–146
Approaching minimum — model capacity saturation
Loss: 7.082 → 6.996Minimum LR, weight decay regularization
Training Metrics
Loss Curve
?
?
?
?
Smoothed Loss
Perplexity
Learning Rate
Gradient Norm
Throughput (tok/s)
Timing Breakdown
No Telemetry
Model Architecture
Model Configuration
ArchitectureGPT (decoder-only transformer)
Parameters55.77M
Layers16
Embedding Dim512
Attention Heads8
Head Dim64
FFN Dim2048
FFN Activationswiglu
Vocab Size4,000
Context Length512 tokens
Dropout0
Training Configuration
Optimizeradamw
Learning Rate0.0006
LR Min0.00006
LR ScheduleCosine decay
Warmup Steps500
Batch Size4
Grad Accum Steps1
Effective Batch4
Grad Clip1
Weight Decay0.1
Backendhelios
Tokenizerbpe-chat-4k
Seed42
Layer Structure
Token Embed
4,000×512
Pos Embed
512×512
Block 0
Attn+FFN
Block 1
Attn+FFN
Block 2
Attn+FFN
Block 3
Attn+FFN
Block 4
Attn+FFN
Block 5
Attn+FFN
LayerNorm
512
LM Head
512×4,000
Generated Samples
No samples generated yet. Samples appear at configured intervals during training.
Checkpoints
No checkpoints saved yet.
Chat with Model
Send a message to chat with this model
Generated Invalid Date Invalid Date — Alpha Training SystemConfig hash: