INDEX

Explanations

words related to sipping or drinking

oai_token-act-pair · gpt-3.5-turbo

gerunds and present participles

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

rules

-0.69

jug

-0.64

lis

-0.63

 Niet

-0.62

sighted

-0.60

 Firearms

-0.60

oola

-0.60

ossier

-0.59

esc

-0.58

las

-0.57

POSITIVE LOGITS

ificant

0.78

terday

0.72

llah

0.64

 SERV

0.62

ucker

0.60

 extraord

0.58

eper

0.58

agogue

0.57

demand

0.57

 PROGRAM

0.57

Activations Density 0.101%

words related to sipping or drinking

gerunds and present participles

No Comments

No Known Activations