INDEX

Explanations

references to previous research studies and their results

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

#+#

-0.70

 שוליים

-0.62

/**

-0.60

ThroughAttribute

-0.57

 HasFactory

-0.54

قایناق‌لار

-0.54

Tembelea

-0.54

Искәрмәләр

-0.53

EndContext

-0.51

 estekak

-0.50

POSITIVE LOGITS

 Référence

0.42

Bibliograf

0.39

ๆ

0.38

KEYCODE

0.36

ัน

0.35

 published

0.34

 artykuł

0.34

reported

0.34

ffilm

0.34

Autoritní

0.33

Activations Density 0.271%

references to previous research studies and their results

No Comments

No Known Activations