ocr_pilot100k
, trained on L4 (AWS, spot). .
Hypothesis
Bulk OCR of about 100,000 scanned pages (127 books) with OCR-VLM v1 on spot GPUs.
What we learned
Runs in short spot sessions that resume page by page.
Scores
No scores published yet. The held-out scores are published when the run finishes; the live card above shows the latest evaluation during training.
Curves
Training loss
Cross-entropy per token, by step. The first few percent of the run, far higher, run off the top; hover or the table has every value.
Held-out bits per byte
The validation split, evaluated during training, by step. Lower is better.
Throughput
Tokens per second, by step.
Model
- Preset
- —
- Parameters
- —
- Outside embeddings
- —
- Layers · heads · width
- — · — · —
- Tokenizer
- —
Data
- Slice
- —
- Words
- —
- Training tokens
- —
- Passes
- —
- Tokens seen
- —
Compute
- GPU
- L4
- Where
- AWS (spot)
- Steps
- —
- GPU hours
- —
- Cost
- $5.81
- Spot restarts
- 0
Lineage
awsspot