INDEX

Explanations

FAQ-related terms indicating help or information resources

oai_token-act-pair · gpt-3.5-turbo

mentions of FAQ sections or related inquiries

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rings

-0.76

rigan

-0.75

onde

-0.75

intent

-0.73

arnaev

-0.72

kus

-0.71

wcs

-0.70

paren

-0.69

heimer

-0.69

right

-0.68

POSITIVE LOGITS

 Questions

0.85

 questions

0.82

naires

0.81

FAQ

0.79

 Answer

0.77

 answered

0.76

 quer

0.75

 quizz

0.73

FAQ

0.68

 Asked

0.68

Activations Density 0.040%

FAQ-related terms indicating help or information resources

mentions of FAQ sections or related inquiries

No Comments

No Known Activations