INDEX

Explanations

locations or areas within a building, specifically rooms

oai_token-act-pair · gpt-3.5-turbo

references to different types of rooms in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 reconnect

-0.68

 Kickstarter

-0.67

 compelling

-0.62

 Governments

-0.61

uras

-0.61

 frivolous

-0.60

 sacr

-0.59

olls

-0.58

 Mother

-0.57

RBI

-0.57

POSITIVE LOGITS

room

1.17

rooms

1.08

idge

0.92

bell

0.87

stairs

0.84

bial

0.83

ichick

0.83

ãĤ£

0.81

itory

0.81

house

0.79

Activations Density 0.009%

locations or areas within a building, specifically rooms

references to different types of rooms in various contexts

No Comments

No Known Activations