INDEX

Explanations

the word "no" followed by a concept or entity

oai_token-act-pair · gpt-3.5-turbo

the repetition of the phrase "no one."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CLOSE

-0.67

ktop

-0.58

 PLUS

-0.57

 Rounds

-0.56

Launch

-0.56

ciation

-0.55

 Plaza

-0.55

edged

-0.55

 Cinema

-0.55

rez

-0.54

POSITIVE LOGITS

 matter

1.25

xious

1.16

one

1.08

 sane

1.04

 amount

1.00

one

0.98

 sooner

0.96

 conceivable

0.94

 doubt

0.93

 longer

0.92

Activations Density 0.057%

the word "no" followed by a concept or entity

the repetition of the phrase "no one."

No Comments

No Known Activations