arm
fn_span_marker_n200_v1 baseline
One config on few_nerd_280, scored one way; its repeats averaged per item.
- dataset
- few_nerd_280
- scored against
- gold/few_nerd_280 gold
- goal
- Named-entity extraction (Few-NERD)
- runs
- 1
- headline
- f1 0.767
- last run
- 2026-09-28 07:08
01metrics
| metric | mean | items | better |
|---|---|---|---|
| f1 | 0.767 | 280 | higher |
| cost_usd | $0.000000 | 280 | lower |
| fn | 0.539 | 280 | neutral |
| fp | 0.593 | 280 | neutral |
| latency_ms | 757.9 | 280 | lower |
| n_gold | 2.736 | 280 | neutral |
| n_predicted | 2.789 | 280 | neutral |
| parsed | 1.000 | 280 | higher |
| precision | 0.773 | 280 | higher |
| recall | 0.774 | 280 | higher |
| tp | 2.196 | 280 | neutral |
| type_penalty | 0.070 | 280 | neutral |
| untyped_f1 | 0.837 | 280 | higher |
| untyped_fn | 0.361 | 280 | neutral |
| untyped_fp | 0.414 | 280 | neutral |
| untyped_tp | 2.375 | 280 | neutral |
02cost and speed
- cost per item
- $0.000000
- speed per item
- p50 630 · p95 1,217 ms
Per item, over its items: cost as dollars and tokens (a local model is free but still uses tokens), speed as the median and the slow tail (p95). A goal that names this arm weighs them in its decision.
03runs
| run | made | items | cost | how |
|---|---|---|---|---|
fn_span_marker_n200_v1_20260928T070856Z | 2026-09-28 07:08 | 280 | $0.000000 | rescore |
04config
config fn_span_marker_n200_v1 · adapter examples/ner-few-nerd/adapter.py · credentials are never shown
- coarsen
true- device
cpu- family
span_marker- model
guishe/span-marker-generic-ner-v1-fewnerd-fine-super- precision
fp32- price_tier
free- provider
span_marker
05deploy sheet
- measured by
fn_span_marker_n200_v1_20260928T070856Zon few_nerd_280- against
- gold/few_nerd_280
- model sent
guishe/span-marker-generic-ner-v1-fewnerd-fine-super- model record
guishe-span-marker-generic-ner-v1-fewnerd-fine-super- server
- not recorded
- build
- f836dfa9b6b0e6a6379607a8e35166103ac41aee
- settings sent
coarsen: truedevice: cpumodel: guishe/span-marker-generic-ner-v1-fewnerd-fine-superprecision: fp32provider: span_marker- not sent
- bookkeeping:
family, price_tier - declared
deployment.coarsen: Truedeployment.labels: ['art', 'building', 'event', 'location', 'organization', 'other', 'person', 'product']deployment.normalizer_sha256: a7c22b054099a72f91d772754dcd49b748fa739340ea4f38f2934eb18950a942deployment.parser_sha256: 101110f889f26b800af7eacbd887bab82aef1556713b869ab27d86f78a605181deployment.provider: span_markerdeployment.weight_files: ['model.safetensors:499028932']- adapter
examples/ner-few-nerd/adapter.pyd147da551073- harness
- b3cc5d861cf2
No capture here says what production ran (the adapter's production_ran).
Command: evh deploy-sheet --config fn_span_marker_n200_v1 [--capture C]
06conditions
- data.items_sha256
95a1b962707368cce0895143e73b63cd9884425502997ee1a93df99f97e2cad7- deployment.build
f836dfa9b6b0e6a6379607a8e35166103ac41aee- deployment.device
cpu- deployment.model
guishe/span-marker-generic-ner-v1-fewnerd-fine-super- deployment.quantization
fp32- deployment.source
adapter- fingerprint.version
5- library.datasets
5.0.1- library.numpy
1.26.4- library.openai
3.19.2- library.scikit-learn
1.9.1- library.torch
2.2.2- library.transformers
4.55.4- machine.cpu_count
12- machine.label
mac-dev- machine.platform
Darwin/x86_64- machine.processor
i386- machine.python
3.11.8- machine.release
24.6.0- reference.sha256
7cdd09fe0084ba29c616ab65fbfa08bf149e9006819641e9e91fe502ebe15a2b- scorer.id
examples/ner-few-nerd/adapter.py- scorer.sha256
d147da55107305786c21b75ac87d748b9ea4fb3e57ebdff622aafecd826559b1
07rounds
- Round r1 of Named-entity extraction (Few-NERD), closed: one of its arms.