INDEX

Explanations

mentions of food-related terms

oai_token-act-pair · gpt-3.5-turbo

references to food and its related issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orp

-0.90

 Freem

-0.75

acular

-0.71

 Masonic

-0.69

 Werewolf

-0.69

inct

-0.67

 Gemini

-0.67

 Vaughn

-0.66

 Wink

-0.66

 Studio

-0.65

POSITIVE LOGITS

stuff

0.98

corn

0.98

 eaten

0.96

stocks

0.93

 nutrit

0.92

grain

0.92

 sacrific

0.88

 grains

0.88

 crops

0.87

food

0.86

Activations Density 0.034%

mentions of food-related terms

references to food and its related issues

No Comments

No Known Activations