INDEX

Explanations

proper names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to people's roles, titles, or positions of authority

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

 trunc

-0.67

SPONSORED

-0.65

Split

-0.63

inosaur

-0.63

Contents

-0.62

 dstg

-0.62

netflix

-0.62

 calendar

-0.62

EXP

-0.62

Tube

-0.61

POSITIVE LOGITS

 spokesman

0.92

PhD

0.92

 spokesperson

0.91

Jr

0.90

 researcher

0.86

 spokeswoman

0.86

 lecturer

0.85

 Managing

0.85

 dean

0.82

 professor

0.82

Activations Density 0.157%

proper names of individuals

references to people's roles, titles, or positions of authority

No Comments

No Known Activations