INDEX

Explanations

instances where the text discusses the implications or consequences of certain actions or decisions

oai_token-act-pair · gpt-3.5-turbo

references to the word "it" often in contexts suggesting discussion of a subject or object

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

UGH

-0.71

OME

-0.69

IFE

-0.66

 Finish

-0.64

Hope

-0.64

FUL

-0.62

Flight

-0.61

Magikarp

-0.61

 Genius

-0.61

hift

-0.61

POSITIVE LOGITS

 involves

1.33

 relates

1.32

 contradicts

1.28

 coincides

1.23

 represents

1.22

 embodies

1.18

 violates

1.18

 contains

1.17

 resembles

1.16

 reflects

1.15

Activations Density 0.216%

instances where the text discusses the implications or consequences of certain actions or decisions

references to the word "it" often in contexts suggesting discussion of a subject or object

No Comments

No Known Activations