INDEX

Explanations

references to academic and research-related terms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

liceerd

-0.43

enumii

-0.42

 Normdatei

-0.42

 <=",

-0.42

InstrumentedTest

-0.41

kie

-0.41

AndEndTag

-0.40

 Tracer

-0.40

 CURI

-0.40



-0.39

POSITIVE LOGITS

 Photocase

0.61

Portail

0.56

Personensuche

0.50

GEBURTSDATUM

0.44

 gynhyrchwyd

0.42

 незавершена

0.41

sizeCache

0.39

mergeFrom

0.39

Portale

0.39

Espèce

0.39

Activations Density 0.197%

references to academic and research-related terms

No Comments

No Known Activations