arm
db_deepseek_l_n200_v1
One config on dbpedia_280, scored one way; its repeats averaged per item.
- dataset
- dbpedia_280
- scored against
- gold/dbpedia_280 gold
- goal
- Encyclopedia topic classification (DBpedia)
- runs
- 1
- headline
- correct 0.986
- last run
- 2026-09-27 14:21
Inspect itAgainst the baselineWith the 29 arms scored the same way
01metrics
| metric | mean | items | better |
|---|---|---|---|
| correct | 0.986 | 280 | higher |
| answer_words | 1.000 | 280 | neutral |
| correct_after_repair | 0.000 | 280 | neutral |
| cost_usd | $0.000074 | 280 | lower |
| fmt_markdown | 0.000 | 280 | neutral |
| fmt_verbose | 0.000 | 280 | neutral |
| latency_ms | 1781.5 | 280 | lower |
| parsed | 1.000 | 280 | higher |
| reasoning_tokens | 0.000 | 280 | neutral |
| tokens_in | 138.7 | 280 | lower |
| tokens_out | 2.636 | 280 | lower |
| truncated | 0.000 | 280 | neutral |
02cost and speed
- cost per item
- $0.000074 · 141 tok
- speed per item
- p50 1,497 · p95 3,617 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 |
|---|---|---|---|---|
db_deepseek_l_n200_v1_20260927T141255Z | 2026-09-27 14:21 | 280 | $0.020855 | measured |
04config
config db_deepseek_l_n200_v1 · adapter examples/classification-dbpedia-14/adapter.py · credentials are never shown
- family
deepseek- max_tokens
200- model
eval-deepseek-pro- price_tier
large- prompt_file
prompt.txt- provider
litellm- reasoning
{"enabled": false}- temperature
0.0- usd_per_mtok_in
0.28- usd_per_mtok_out
1.1
05conditions
- data.items_sha256
a7419dc6a5876490c3ed1f96eae9398969a2f8b794e744d45ace5790bc0146f2- deployment.model
eval-deepseek-pro- deployment.source
adapter- deployment.upstream
openrouter/deepseek/deepseek-v4-pro- fingerprint.version
5- library.openai
3.19.2- 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
e0564fe65fb011e37717e037a6ad662f44e796b7d7de8e62ed2344ba56fa7a8a- scorer.id
examples/classification-dbpedia-14/adapter.py- scorer.sha256
58b55c964f8bb40746c359621beb25bfd0a071fca079961bdf15be1e7c0a77a8
06rounds
- Round r1 of Encyclopedia topic classification (DBpedia), closed: one of its arms.