INDEX

Explanations

phrases with the word "more" followed by a number indicating a quantity

oai_token-act-pair · gpt-3.5-turbo

instances of the word "more" indicating an increase or continuation in various contexts

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

xtap

-0.75

pex

-0.62

ress

-0.62

 Everywhere

-0.60

FORE

-0.60

ĺħ

-0.59

 Summoner

-0.59

 Citation

-0.59

 Difference

-0.57

 Observer

-0.57

POSITIVE LOGITS

 than

1.69

than

1.34

 Than

1.02

ado

0.84

 stringent

0.79

HUD

0.77

 importantly

0.72

 ambitious

0.71

 recently

0.70

ened

0.69

Activations Density 0.091%

phrases with the word "more" followed by a number indicating a quantity

instances of the word "more" indicating an increase or continuation in various contexts

No Comments

No Known Activations