INDEX

Explanations

performance-related metrics and comparisons

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

high and low degrees

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

UserScript

-0.63

 ſua

-0.61

faster

-0.60

 ſuo

-0.60

 greener

-0.57

EditorBrowsable

-0.57

TestingModule

-0.56

 كومونز

-0.56

ंदीखरीदारी

-0.56

RTEX

-0.56

POSITIVE LOGITS

 enough

0.50

 high

0.40

 good

0.39

 genug

0.39

 hohe

0.38

enough

0.37

 Capacidad

0.37

 genoeg

0.36

 strong

0.35

 large

0.34

Activations Density 0.162%

performance-related metrics and comparisons

high and low degrees

No Comments

No Known Activations