Hugging Face's logo

gnumanth
/
nullspan

Question Answering
Transformers
ONNX
Safetensors
nullspan
feature-extraction
system-one
typed-decisions
jevbench
structured-extraction
span-extraction
calibrated-estimation
electra
modernbert
custom_code
Eval Results
Model card Files
xet
 Community
1
gnumanth commited on
Commit
ba74fb2
路
verified 路
1 Parent(s): d4e2679

Add Typed Decisions eval results (v1.2.1)

Browse files
Files changed (1) hide show
  1. .eval_results/typed-decisions.yaml +4 -0
.eval_results/typed-decisions.yaml CHANGED
@@ -6,6 +6,7 @@
6
  source:
7
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
8
  name: "Typed Decisions"
 
9
  notes: "Pure non-LLM bidirectional cross-encoder (DeBERTa-v3-Large 435M + LoRA + Cross-Option & Cross-Question Workflow Competition Head, 0 generated output tokens). Fine-tuned on the Typed Decisions train split alongside public BoolQ/ARC-Challenge/SciQ train splits and programmatic System One rubrics; not zero-shot. Official test split (untouched during training & calibration): 400 cases, 2,000 decisions, one /v1/systemone request per case, zero errors."
10
  - dataset:
11
  id: LocalLLaMA/typed-decisions
@@ -15,6 +16,7 @@
15
  source:
16
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
17
  name: "Typed Decisions"
 
18
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."
19
  - dataset:
20
  id: LocalLLaMA/typed-decisions
@@ -24,6 +26,7 @@
24
  source:
25
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
26
  name: "Typed Decisions"
 
27
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."
28
  - dataset:
29
  id: LocalLLaMA/typed-decisions
@@ -33,4 +36,5 @@
33
  source:
34
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
35
  name: "Typed Decisions"
 
36
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."
 
6
  source:
7
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
8
  name: "Typed Decisions"
9
+ user: gnumanth
10
  notes: "Pure non-LLM bidirectional cross-encoder (DeBERTa-v3-Large 435M + LoRA + Cross-Option & Cross-Question Workflow Competition Head, 0 generated output tokens). Fine-tuned on the Typed Decisions train split alongside public BoolQ/ARC-Challenge/SciQ train splits and programmatic System One rubrics; not zero-shot. Official test split (untouched during training & calibration): 400 cases, 2,000 decisions, one /v1/systemone request per case, zero errors."
11
  - dataset:
12
  id: LocalLLaMA/typed-decisions
 
16
  source:
17
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
18
  name: "Typed Decisions"
19
+ user: gnumanth
20
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."
21
  - dataset:
22
  id: LocalLLaMA/typed-decisions
 
26
  source:
27
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
28
  name: "Typed Decisions"
29
+ user: gnumanth
30
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."
31
  - dataset:
32
  id: LocalLLaMA/typed-decisions
 
36
  source:
37
  url: https://huggingface.co/datasets/LocalLLaMA/typed-decisions
38
  name: "Typed Decisions"
39
+ user: gnumanth
40
  notes: "Pure non-LLM bidirectional cross-encoder (0 generated output tokens). Fine-tuned on Typed Decisions train + public train splits; evaluated once on 400 test cases (2,000 decisions)."