INDEX

Explanations

phrases indicating opposition or conflict

oai_token-act-pair · gpt-3.5-turbo

the word "against" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

overed

-0.68

 Highlands

-0.63

lder

-0.60

 Mystery

-0.56

 Sunrise

-0.56

mitter

-0.55

 cues

-0.54

ffe

-0.53

esthes

-0.52

 fanc

-0.52

POSITIVE LOGITS

 against

3.50

against

3.39

 Against

2.65

Against

2.48

AGA

1.48

 versus

1.37

vs

1.24

 opposed

1.20

 defensively

0.99

 opposing

0.98

Activations Density 0.073%

phrases indicating opposition or conflict

the word "against" in various contexts

No Comments

No Known Activations