INDEX

Explanations

numerical lists of items or tips

oai_token-act-pair · gpt-3.5-turbo

numerical items or lists indicating rankings and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ierrez

-0.67

istration

-0.65

adra

-0.64

drawn

-0.64

 remission

-0.63

administ

-0.63

 attendance

-0.63

ufact

-0.63

verson

-0.62

externalActionCode

-0.61

POSITIVE LOGITS

 Reasons

1.42

 Things

1.24

 Ways

1.21

 Greatest

1.05

 reasons

1.02

 Worst

1.01

 Favorite

0.98

 myths

0.97

 Questions

0.97

 Interesting

0.95

Activations Density 0.103%

numerical lists of items or tips

numerical items or lists indicating rankings and recommendations

No Comments

No Known Activations