INDEX

Explanations

specific references to office spaces

oai_token-act-pair · gpt-3.5-turbo

references to physical locations or entities related to offices

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

oise

-0.88

âķĲâķĲ

-0.80

MQ

-0.75

theless

-0.71

Movie

-0.70

 Haram

-0.67

CHAT

-0.67

ICAN

-0.67

Interstitial

-0.67

ivals

-0.65

POSITIVE LOGITS

holder

1.19

holders

1.18

tops

1.03

room

0.90

 holder

0.86

 prosec

0.83

mate

0.81

person

0.79

cle

0.77

 holders

0.76

Activations Density 0.030%

specific references to office spaces

references to physical locations or entities related to offices

No Comments

No Known Activations