INDEX

Explanations

questions or uncertain statements

oai_token-act-pair · gpt-3.5-turbo

instances of the word "question."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rites

-0.90

ufact

-0.88

emetery

-0.87

orpor

-0.77

alty

-0.76

é¾

-0.72

 satell

-0.72

azeera

-0.72

rylic

-0.71

tsky

-0.70

POSITIVE LOGITS

naires

1.63

naire

1.36

 unanswered

0.92

 posed

0.89

 question

0.86

 mark

0.85

 questions

0.83

 answered

0.78

 asked

0.78

Ans

0.77

Activations Density 0.034%

questions or uncertain statements

instances of the word "question."

No Comments

No Known Activations