INDEX

Explanations

references to the object itself within a sentence

oai_token-act-pair · gpt-3.5-turbo

the word "itself" and its variations

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

ammy

-0.74

onal

-0.68

 Wire

-0.65

 Derby

-0.61

Ern

-0.61

 coworkers

-0.59

oute

-0.58

odies

-0.58

 Kimber

-0.58

 Federation

-0.57

POSITIVE LOGITS

é¾įåĸļå£«

0.84

 contained

0.72

 contains

0.69

 acknowledges

0.69

contained

0.69

 admits

0.66

ainer

0.65

ortium

0.64

 constitutes

0.64

 contradicts

0.62

Activations Density 0.051%

references to the object itself within a sentence

the word "itself" and its variations

No Comments

No Known Activations