INDEX

Explanations

the word "expression."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "expression" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Shades

-0.81

 Brewers

-0.78

sonian

-0.78

 Claus

-0.72

 Bullets

-0.69

elcome

-0.69

END

-0.68

ether

-0.67

Peb

-0.65

Ń·

-0.65

POSITIVE LOGITS

 expression

0.99

terday

0.95

 expressions

0.85

ively

0.82

 Expression

0.80

 guiActiveUn

0.78

 ende

0.77

naire

0.76

pieces

0.76

ist

0.76

Activations Density 0.013%

the word "expression."

instances of the word "expression" and its various contexts

No Comments

No Known Activations