INDEX

Explanations

quantitative data or statistics presented in a visual format, like graphs or charts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating statistical data and trends

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

???

-0.75

tumblr

-0.69

nir

-0.68

Lago

-0.67

Engine

-0.66

bara

-0.65

venture

-0.64

thood

-0.64

another

-0.64

KER

-0.64

POSITIVE LOGITS

 approximate

1.23

 entirety

1.18

 number

1.16

 amount

1.15

 extent

1.14

 entire

1.14

 contents

1.12

 percentage

1.11

 probabilities

1.10

 exact

1.10

Activations Density 0.281%

quantitative data or statistics presented in a visual format, like graphs or charts

phrases indicating statistical data and trends

No Comments

No Known Activations