INDEX

Explanations

expressions of concern or worry

oai_token-act-pair · gpt-3.5-turbo

phrases indicating concerns or worries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

llular

-0.81

NetMessage

-0.70

ucle

-0.68

odes

-0.67

clerosis

-0.65

iens

-0.64

oiler

-0.63

ops

-0.63

format

-0.63

 Brus

-0.63

POSITIVE LOGITS

enance

0.75

 respecting

0.69

 prick

0.68

 Boko

0.66

 vehemently

0.64

Parents

0.60

 Lank

0.59

 homophobic

0.59

unity

0.59

 DEFENSE

0.58

Activations Density 0.329%

expressions of concern or worry

phrases indicating concerns or worries

No Comments

No Known Activations