INDEX

Explanations

evaluations of experiences, focusing on positive and negative descriptors of things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

URLException

-0.57

 SafeMath

-0.48

 restTemplate

-0.48

lise

-0.46

 TimeUnit

-0.46

minton

-0.46

ANTLR

-0.46

existsSync

-0.43

 gluon

-0.43

*/,

-0.43

POSITIVE LOGITS

 about

2.05

about

1.56

 ABOUT

1.35

About

1.33

 About

1.31

 tentang

1.24

ABOUT

1.07

bout

0.95

 aspect

0.94

abt

0.93

Activations Density 0.136%

evaluations of experiences, focusing on positive and negative descriptors of things

No Comments

No Known Activations