INDEX

Explanations

phrases or words related to the final stages or outcomes of a process

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "final" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

holes

-0.69

shire

-0.65

kun

-0.64

 Spears

-0.64

LI

-0.64

 Everywhere

-0.63

stadt

-0.63

heid

-0.62

relative

-0.62

men

-0.62

POSITIVE LOGITS

 FANTASY

0.85

 installment

0.82

 showdown

0.81

 resting

0.81

iminary

0.79

ebin

0.79

 curtain

0.78

 exam

0.75

 countdown

0.74

izers

0.73

Activations Density 0.022%

phrases or words related to the final stages or outcomes of a process

occurrences of the word "final" in various contexts

No Comments

No Known Activations