INDEX

Explanations

the word "are" repeated multiple times

oai_token-act-pair · gpt-3.5-turbo

the word "are" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

OOL

-0.61

Nev

-0.60

uration

-0.60

erguson

-0.59

 shape

-0.58

omez

-0.58

allery

-0.58

inosaur

-0.58

ulates

-0.58

ues

-0.57

POSITIVE LOGITS

nce

1.05

nces

1.00

tsky

0.90

nes

0.85

than

0.84

nt

0.83

atra

0.82

edia

0.81

lli

0.81

tto

0.81

Activations Density 0.014%

the word "are" repeated multiple times

the word "are" in various contexts

No Comments

No Known Activations