stale
super_chat_20260310074201_vxnn
97.65M parameter unknown model — bpe-8k-chat tokenizer, 12L/768D/12H
Overview
97.65M
Parameters
5.0352
Final Loss
5.0464
Best Val Loss
153.7
Perplexity
18,534,400
Tokens Processed
0.2
Tokens/Param
2,042 tok/s
Avg Throughput
10h 5m
Training Time
Training Progress4,525 / 6,000 steps (75.4%)
Loss reduced by 44.9% from initial 9.1337
Dataset & Training
Domainunknown
Tokenizerbpe-8k-chat
Total Iterations6,000
Batch Size8
Context Length512 tokens
Tokens per Batch4,096
Dataset Passes~23
Effective Tokens18,534,400
Training Pipeline
Warmupsteps 1–60
Learning rate warmup — model weights adjusting to data distribution
Loss: 9.134 → 7.381Linear LR warmup, gradient clipping
Rapid Descentsteps 60–1,801
Steepest loss reduction — model learning primary patterns
Loss: 7.381 → 5.294Cosine LR schedule, AdamW optimization
Refinementsteps 1,801–4,201
Diminishing returns — model fine-tuning subtler patterns
Loss: 5.294 → 5.111Lower LR, gradient accumulation
Convergencesteps 4,201–4,525
Approaching minimum — model capacity saturation
Loss: 5.111 → 5.035Minimum 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)
Parameters97.65M
Layers12
Embedding Dim768
Attention Heads12
Head Dim64
FFN Dim3072
FFN Activationgelu
Vocab Size8,000
Context Length512 tokens
Dropout0
Training Configuration
Optimizeradamw
Learning Rate0.0003
LR Min0.00003
LR ScheduleCosine decay
Warmup Steps500
Batch Size8
Grad Accum Steps4
Effective Batch32
Grad Clip1
Weight Decay0.1
Backendhelios
Tokenizerbpe-8k-chat
Seed42
Layer Structure
Token Embed
8,000×768
Pos Embed
512×768
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
768
LM Head
768×8,000
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: