INDEX

Explanations

phrases related to actions or events involving others without their consent

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases indicating alternatives or inclusivity

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rehensive

-0.69

NC

-0.68

othal

-0.67

geist

-0.66

rier

-0.66

CIS

-0.61

ennial

-0.61

uble

-0.61

Found

-0.61

WM

-0.59

POSITIVE LOGITS

 inhal

0.75

 conscience

0.73

 adorned

0.69

 unheard

0.68

 without

0.68

 vulner

0.67

 manner

0.67

 devoid

0.66

ebook

0.65

 notation

0.65

Activations Density 0.500%

phrases related to actions or events involving others without their consent

conjunctions and phrases indicating alternatives or inclusivity

No Comments

No Known Activations