INDEX

Explanations

phrases related to responding aggressively or defensively in an argument or conflict

oai_token-act-pair · gpt-3.5-turbo

instances of confrontational responses or criticisms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ocument

-0.77

 Transform

-0.77

minster

-0.72

ancial

-0.71

scope

-0.70

utical

-0.68

benef

-0.66

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

soDeliveryDate

-0.64

inct

-0.64

POSITIVE LOGITS

 accusing

0.91

 criticism

0.85

 questioning

0.84

 critics

0.84

 angrily

0.78

 blaming

0.78

 criticisms

0.77

 mockery

0.76

 commenters

0.75

 critic

0.75

Activations Density 0.259%

phrases related to responding aggressively or defensively in an argument or conflict

instances of confrontational responses or criticisms

No Comments

No Known Activations