INDEX

Explanations

phrases containing the word "are" along with various other words

oai_token-act-pair · gpt-3.5-turbo

the word "are" 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 @ 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

âĶ

-0.60

âĢİ

-0.60

odder

-0.60

irmation

-0.56

é¾

-0.54

icism

-0.51

ique

-0.51

 Challenger

-0.51

 McKin

-0.50

ication

-0.48

POSITIVE LOGITS

senal

1.31

wolves

1.18

wolf

0.99

types

0.78

 trademarks

0.74

 embodiments

0.71

pas

0.70

 SHARES

0.66

 examples

0.66

 sinners

0.63

Activations Density 0.050%

phrases containing the word "are" along with various other words

the word "are" in various contexts

No Comments

No Known Activations