yafitzdev commited on
Commit
e4ec577
·
verified ·
1 Parent(s): 4b66447

Update v1 model card names

Browse files
Files changed (1) hide show
  1. README.md +8 -6
README.md CHANGED
@@ -19,9 +19,11 @@ metrics:
19
  - f1
20
  ---
21
 
22
- # pyrrho-nano-g2
 
 
23
 
24
- pyrrho-nano-g2 is a small RAG governance co-processor for anti-hallucination pipelines. It reads a user question plus retrieved source passages and returns an evidence-state decision a RAG application can use before answering: `ABSTAIN`, `DISPUTED`, or `TRUSTWORTHY`.
25
 
26
  It is not an answer generator and not an open-world fact checker. It sits between retrieval and generation, or beside a generator as a fast guardrail, to reduce cases where unsupported or contradictory retrieved evidence gets treated as safe to answer from.
27
 
@@ -95,7 +97,7 @@ This model is not intended to write answers, verify facts outside the provided s
95
  from transformers import AutoModelForSequenceClassification, AutoTokenizer
96
  import torch
97
 
98
- MODEL_ID = "yafitzdev/pyrrho-nano-g2"
99
  LABELS = ["ABSTAIN", "DISPUTED", "TRUSTWORTHY"]
100
  TAU = 0.50
101
 
@@ -150,7 +152,7 @@ from transformers import AutoTokenizer
150
  import numpy as np
151
  import onnxruntime as ort
152
 
153
- MODEL_ID = "yafitzdev/pyrrho-nano-g2"
154
  query = "Has the company achieved profitability?"
155
  contexts = [
156
  "The company posted its first profitable quarter, with net income of $4 million.",
@@ -263,10 +265,10 @@ The validation split was used for checkpoint and threshold selection. The held-o
263
 
264
  ```bibtex
265
  @misc{pyrrho_nano_g2_2026,
266
- title = {pyrrho-nano-g2},
267
  author = {Yan Fitzner},
268
  year = {2026},
269
- url = {https://huggingface.co/yafitzdev/pyrrho-nano-g2},
270
  }
271
  ```
272
 
 
19
  - f1
20
  ---
21
 
22
+ > Historical Pyrrho v1 release. Current v2 model: [yafitzdev/pyrrho-v2-nano-g1](https://huggingface.co/yafitzdev/pyrrho-v2-nano-g1).
23
+
24
+ # pyrrho-v1-nano-g2
25
 
26
+ pyrrho-v1-nano-g2 is a small RAG governance co-processor for anti-hallucination pipelines. It reads a user question plus retrieved source passages and returns an evidence-state decision a RAG application can use before answering: `ABSTAIN`, `DISPUTED`, or `TRUSTWORTHY`.
27
 
28
  It is not an answer generator and not an open-world fact checker. It sits between retrieval and generation, or beside a generator as a fast guardrail, to reduce cases where unsupported or contradictory retrieved evidence gets treated as safe to answer from.
29
 
 
97
  from transformers import AutoModelForSequenceClassification, AutoTokenizer
98
  import torch
99
 
100
+ MODEL_ID = "yafitzdev/pyrrho-v1-nano-g2"
101
  LABELS = ["ABSTAIN", "DISPUTED", "TRUSTWORTHY"]
102
  TAU = 0.50
103
 
 
152
  import numpy as np
153
  import onnxruntime as ort
154
 
155
+ MODEL_ID = "yafitzdev/pyrrho-v1-nano-g2"
156
  query = "Has the company achieved profitability?"
157
  contexts = [
158
  "The company posted its first profitable quarter, with net income of $4 million.",
 
265
 
266
  ```bibtex
267
  @misc{pyrrho_nano_g2_2026,
268
+ title = {pyrrho-v1-nano-g2},
269
  author = {Yan Fitzner},
270
  year = {2026},
271
+ url = {https://huggingface.co/yafitzdev/pyrrho-v1-nano-g2},
272
  }
273
  ```
274