INDEX

Explanations

phrases related to being in a specific location or situation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" followed by varying frequencies to indicate locations or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 incumb

-0.70

NOW

-0.69

named

-0.66

llor

-0.65

caster

-0.65

 ende

-0.65

 debuted

-0.64

linger

-0.64

ZA

-0.63

killed

-0.63

POSITIVE LOGITS

 lieu

1.20

 accordance

1.16

 spite

1.02

 vitro

1.01

animate

0.99

 vain

0.99

ordinate

0.96

clus

0.93

 order

0.93

escap

0.91

Activations Density 0.390%

phrases related to being in a specific location or situation

occurrences of the word "in" followed by varying frequencies to indicate locations or states

No Comments

No Known Activations