INDEX

Explanations

instances of the word "couple" followed by a numerical value

oai_token-act-pair · gpt-3.5-turbo

the term "couple" used 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 @ 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

schild

-0.77

ulhu

-0.75

kinson

-0.74

MpServer

-0.69

é¾

-0.68

 Directorate

-0.67

ibaba

-0.66

farious

-0.63

utherford

-0.63

ç¥ŀ

-0.62

POSITIVE LOGITS

 dozen

0.95

 hundred

0.85

ples

0.78

zon

0.75

dozen

0.75

tones

0.74

 couple

0.72

Timeout

0.71

tons

0.71

plates

0.69

Activations Density 0.016%

instances of the word "couple" followed by a numerical value

the term "couple" used in various contexts

No Comments

No Known Activations