INDEX

Explanations

the suffix "-or" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "or"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 msec

-0.65

inctions

-0.63

eas

-0.62

Ĥİ

-0.61

assadors

-0.57

Alert

-0.57

IDENT

-0.55

 Ride

-0.55

lished

-0.55

tnc

-0.55

POSITIVE LOGITS

chid

1.11

ific

1.09

onto

1.05

andom

1.04

phans

1.03

thodox

1.00

orate

0.98

Else

0.96

acle

0.96

ikawa

0.94

Activations Density 0.031%

the suffix "-or" in various contexts

instances of the word "or"

No Comments

No Known Activations