INDEX

Explanations

proper nouns and technical terms related to health, science, and geography

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 ModelExpression

-0.73

RenderAtEndOf

-0.54

CppMethod

-0.52

Бахар

-0.50

ParallelGroup

-0.48

msgTypes

-0.48

Попис

-0.46

GOTREF

-0.45

 ostavi

-0.45

帖最后由

-0.45

POSITIVE LOGITS

que

0.43

 виправивши

0.40

 հղումներ

0.39

igkeit

0.38

IntoConstraints

0.37

AISSEE

0.36

Dislikes

0.35

 NUKAT

0.35

 though

0.35

aronder

0.35

Activations Density 0.039%

proper nouns and technical terms related to health, science, and geography

No Comments

No Known Activations