INDEX

Explanations

phrases related to opting out or opt-out clauses

oai_token-act-pair · gpt-3.5-turbo

phrases indicating options or choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

issance

-0.70

Lau

-0.67

 Gazette

-0.62

Mek

-0.59

 Situation

-0.59

 Helic

-0.59

 Tape

-0.59

enegger

-0.58

 sinks

-0.58

 Problem

-0.58

POSITIVE LOGITS

out

0.92

cert

0.85

posted

0.85

elected

0.85

level

0.85

exit

0.84

mode

0.84

eyed

0.84

present

0.84

ups

0.81

Activations Density 0.034%

phrases related to opting out or opt-out clauses

phrases indicating options or choices

No Comments

No Known Activations