INDEX

Explanations

questions questioning the validity or morality of a situation or action

oai_token-act-pair · gpt-3.5-turbo

phrases that question possibilities and confrontations regarding events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Board

-0.67

hig

-0.65

 curv

-0.63

DAQ

-0.63

RAFT

-0.62

 vert

-0.61

 seiz

-0.61

arella

-0.58

aukee

-0.57

beard

-0.56

POSITIVE LOGITS

berra

1.08

't

1.00

 anyone

0.78

gow

0.73

ADA

0.72

 anybody

0.72

adian

0.72

 bystanders

0.72

aries

0.67

 onlook

0.67

Activations Density 0.048%

questions questioning the validity or morality of a situation or action

phrases that question possibilities and confrontations regarding events or situations

No Comments

No Known Activations