INDEX

Explanations

words related to quantification

oai_token-act-pair · gpt-3.5-turbo

words that indicate universality or inclusivity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

grave

-0.83

arella

-0.72

pload

-0.71

APD

-0.69

leys

-0.68

ertodd

-0.65

etsk

-0.64

anwhile

-0.62

 ende

-0.61

ItemTracker

-0.58

POSITIVE LOGITS

il

0.67

thood

0.62

 defunct

0.61

 impending

0.59

 Ancients

0.59

oneliness

0.58

 Century

0.57

 sorts

0.56

ribe

0.56

ulz

0.56

Activations Density 0.326%

words related to quantification

words that indicate universality or inclusivity

No Comments

No Known Activations