INDEX

Explanations

locations or facilities related to health and community services

oai_token-act-pair · gpt-3.5-turbo

references to various types of centers and clinics

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

*/(

-0.79

ãĥĥãĥī

-0.76

ãģĦ

-0.74

cloud

-0.71

 Verse

-0.67

 Incarn

-0.64

 Rounds

-0.62

latest

-0.62

Dust

-0.62

 recoil

-0.61

POSITIVE LOGITS

 specializing

0.96

 operator

0.93

 operated

0.92

 staffed

0.92

 operators

0.86

yrinth

0.84

orer

0.76

rador

0.75

 catering

0.75

 facilities

0.73

Activations Density 0.313%

locations or facilities related to health and community services

references to various types of centers and clinics

No Comments

No Known Activations