INDEX

Explanations

phrases related to questioning or doubts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "question" and its variations, indicating inquiries or challenges

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

oiler

-0.85

emetery

-0.81

opl

-0.73

ategory

-0.71

uscript

-0.71

rites

-0.70

ufact

-0.70

ergic

-0.70

aghetti

-0.69

ophe

-0.69

POSITIVE LOGITS

naires

1.09

ively

0.92

 questioning

0.90

naire

0.84

 questioned

0.80

 questions

0.79

 whether

0.78

 unanswered

0.75

him

0.72

 doubts

0.70

Activations Density 0.039%

phrases related to questioning or doubts

instances of the word "question" and its variations, indicating inquiries or challenges

No Comments

No Known Activations