INDEX

Explanations

focus on research-related terms and activities

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

"research" and its related forms

np_acts-logits-general · gemini-2.0-flash

research topics and areas

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 duide

-0.50

 amnio

-0.45

 TestBed

-0.45

Simplicity

-0.44

 opioid

-0.44

 nemlig

-0.44

toFloat

-0.44

Mandy

-0.44

 mitral

-0.43

diode

-0.43

POSITIVE LOGITS

 research

2.13

research

1.95

 Research

1.91

 RESEARCH

1.84

Research

1.80

RESEARCH

1.71

 researches

1.50

 researching

1.47

 investigación

1.39

 Forschung

1.38

Activations Density 0.077%

focus on research-related terms and activities

"research" and its related forms

research topics and areas

No Comments

No Known Activations