INDEX

Explanations

studies and reports conducted by various organizations

oai_token-act-pair · gpt-3.5-turbo

references to studies and their authors

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

Interstitial

-0.90

ãĤ¦ãĤ¹

-0.75

reci

-0.73

alogue

-0.72

tone

-0.71

 sidx

-0.71

tub

-0.71

Ì

-0.70

Contents

-0.69

usercontent

-0.69

POSITIVE LOGITS

 researchers

0.98

 consultancy

0.89

 researcher

0.88

 Physicians

0.87

 Bloomberg

0.87

 Johns

0.86

 Univ

0.85

 University

0.84

 economists

0.84

 Princeton

0.81

Activations Density 0.140%

studies and reports conducted by various organizations

references to studies and their authors

No Comments

No Known Activations