INDEX

Explanations

expressions of ability or capability

oai_token-act-pair · gpt-3.5-turbo

phrases indicating capability or ability

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

IFrame

Link

Negative Logits

 Federation

-0.70

 Rising

-0.67

soever

-0.66

quartered

-0.65

 vacated

-0.62

 Hunters

-0.60

bats

-0.58

rant

-0.57

 revision

-0.57

 revisions

-0.56

POSITIVE LOGITS

't

1.83

berra

1.19

NOT

1.18

adian

1.10

 afford

1.04

isters

0.92

tera

0.92

 tell

0.91

 easily

0.89

 attest

0.89

Activations Density 0.159%

expressions of ability or capability

phrases indicating capability or ability

No Known Activations