INDEX

Explanations

phrases related to specific actions or behaviors

oai_token-act-pair · gpt-3.5-turbo

instances of the preposition "in" indicating a location or context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

username

-0.75

CLASSIFIED

-0.72

tal

-0.70

tre

-0.70

antha

-0.67

ibur

-0.64

tar

-0.64

affer

-0.63

linger

-0.63

ÑĮ

-0.63

POSITIVE LOGITS

animate

1.19

 lieu

1.18

 unison

1.16

 accordance

1.09

 conjunction

1.05

 vain

1.04

humane

0.99

 spite

0.99

 bathrooms

0.99

 vitro

0.99

Activations Density 0.274%

phrases related to specific actions or behaviors

instances of the preposition "in" indicating a location or context

No Comments

No Known Activations