INDEX

Explanations

instances of the word "gap" in different contexts

oai_token-act-pair · gpt-3.5-turbo

references to various types of gaps in context, such as social, economic, or informational disparities

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

vic

-0.67

vez

-0.62

ongyang

-0.59

der

-0.59

rpm

-0.58

 deriv

-0.57

Die

-0.57

 Plant

-0.56

 butterflies

-0.56

Eug

-0.56

POSITIVE LOGITS

 between

1.27

between

1.15

 Between

1.01

gap

1.00

wid

0.99

 separating

0.98

 widen

0.97

 widened

0.93

 widening

0.93

 gaps

0.88

Activations Density 0.063%

instances of the word "gap" in different contexts

references to various types of gaps in context, such as social, economic, or informational disparities

No Comments

No Known Activations