INDEX

Explanations

phrases indicating desire or permission to act freely without restriction

oai_token-act-pair · gpt-3.5-turbo

expressions of desire or willingness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

livious

-0.76

 enthusi

-0.71

cling

-0.67

ynski

-0.64

onut

-0.61

BuyableInstoreAndOnline

-0.59

 Echoes

-0.58

pher

-0.57

gre

-0.57

 faint

-0.56

POSITIVE LOGITS

PLA

0.74

 without

0.74

mares

0.72

 WITHOUT

0.69

to

0.69

 regardless

0.69

ably

0.67

 them

0.66

0.65

*.

0.65

Activations Density 0.052%

phrases indicating desire or permission to act freely without restriction

expressions of desire or willingness

No Comments

No Known Activations