INDEX

Explanations

references to different types of research studies (RS) or similar terminologies in scientific contexts

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

rs followed by numbers

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

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

 McIntyre

-0.57

ParallelGroup

-0.56

 Delaney

-0.52

Neighborhood

-0.50

Continent

-0.49

Plaintiff

-0.48

 Immaculate

-0.48

Analyze

-0.47

 Kinney

-0.47

 configureStore

-0.47

POSITIVE LOGITS

RS

1.13

rs

1.13

RS

1.08

rs

1.04

Rs

0.90

Rs

0.75

ros

0.68

srs

0.67

Ers

0.67

SRS

0.66

Activations Density 0.019%

references to different types of research studies (RS) or similar terminologies in scientific contexts

rs followed by numbers

No Comments

No Known Activations