INDEX

Explanations

names of people or entities

oai_token-act-pair · gpt-3.5-turbo

the word "likes" 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 @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ŃĶ

-0.63

 RELEASE

-0.60

INTON

-0.60

ccording

-0.60

 exting

-0.57

 unden

-0.56

gins

-0.55

english

-0.54

 nonpartisan

-0.54

INA

-0.53

POSITIVE LOGITS

of

1.05

liest

1.05

paces

0.96

wikipedia

0.87

lihood

0.79

Of

0.79

 thereof

0.76

creen

0.74

ãĤ¯

0.73

mith

0.73

Activations Density 0.024%

names of people or entities

the word "likes" and its variations

No Comments

No Known Activations