INDEX

Explanations

keywords related to physical locations or real estate

oai_token-act-pair · gpt-3.5-turbo

mentions of "premise" and "premises."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

amer

-0.83

 Tracker

-0.72

witch

-0.66

roller

-0.64

 Gabriel

-0.63

amed

-0.62

 rebound

-0.62

 blur

-0.62

icon

-0.61

 wrench

-0.61

POSITIVE LOGITS

 premises

4.03

 premise

1.90

 Prem

1.37

 establishments

1.28

 grounds

1.27

 precincts

1.16

 faculties

1.06

 principals

1.05

 residences

1.05

 campuses

1.02

Activations Density 0.022%

keywords related to physical locations or real estate

mentions of "premise" and "premises."

No Comments

No Known Activations