INDEX

Explanations

names of places

oai_token-act-pair · gpt-3.5-turbo

the repetitive use of the word "are."

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

ured

-0.70

insula

-0.67

urers

-0.65

omez

-0.65

ingen

-0.65

ulators

-0.63

uring

-0.61

isively

-0.59

shire

-0.58

BD

-0.57

POSITIVE LOGITS

lli

1.36

tto

1.32

tta

1.30

nce

1.18

tsky

1.17

ndra

1.12

llo

1.11

nces

1.09

tti

1.09

tt

1.08

Activations Density 0.048%

names of places

the repetitive use of the word "are."

No Comments

No Known Activations