INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names related to people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

è¦ļéĨĴ

-0.72

atically

-0.70

lund

-0.68

 confir

-0.65

 Adren

-0.64

 fracturing

-0.64

ASE

-0.61

aneously

-0.60

 GEAR

-0.60

holm

-0.60

POSITIVE LOGITS

uve

0.76

llor

0.75

igi

0.75

orno

0.73

vol

0.73

orce

0.72

amn

0.71

yssey

0.70

cius

0.70

opoly

0.69

Activations Density 0.027%

names of specific individuals

proper nouns, particularly names related to people

No Comments

No Known Activations