INDEX

Explanations

questions starting with "Why is" or "Why are" followed by a statement

oai_token-act-pair · gpt-3.5-turbo

questions or statements that utilize the word "why."

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

fman

-0.74

inav

-0.72

 sails

-0.67

bis

-0.66

ares

-0.65

orge

-0.65

 Islands

-0.63

ioxide

-0.63

rette

-0.61

 equivalents

-0.60

POSITIVE LOGITS

?]

0.92

0.63

 intrusive

0.63

>[

0.62

ogene

0.61

 angry

0.61

NPR

0.60

 surprised

0.60

!?

0.60

 alarmed

0.59

Activations Density 0.036%

questions starting with "Why is" or "Why are" followed by a statement

questions or statements that utilize the word "why."

No Comments

No Known Activations