INDEX

Explanations

the word "or" followed by another word or number

oai_token-act-pair · gpt-3.5-turbo

the word "or" in various contexts

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

onday

-0.68

Alert

-0.67

elong

-0.66

ilton

-0.65

opian

-0.64

edom

-0.64

ires

-0.64

 horizont

-0.63

pter

-0.61

emouth

-0.60

POSITIVE LOGITS

chard

1.32

acle

1.23

acles

1.22

Else

1.20

lando

1.12

chid

1.11

acular

1.04

 otherwise

1.01

nam

1.00

ifice

0.97

Activations Density 0.126%

the word "or" followed by another word or number

the word "or" in various contexts

No Comments

No Known Activations