INDEX

Explanations

references to awards or titles

oai_token-act-pair · gpt-3.5-turbo

references to the word "the" in relation to various subjects or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Downloadha

-0.72

SPONSORED

-0.71

fulness

-0.69

ATURES

-0.67

plets

-0.66

ledge

-0.66

,''

-0.65

cases

-0.65

ARA

-0.63

Shut

-0.63

POSITIVE LOGITS

 utmost

1.11

 same

1.07

 highest

1.02

 Confederacy

0.93

 Americas

0.90

 Month

0.88

same

0.88

 apocalypse

0.86

 latter

0.86

era

0.85

Activations Density 0.181%

references to awards or titles

references to the word "the" in relation to various subjects or entities

No Comments

No Known Activations