INDEX

Explanations

specific names and identifiers related to structured data or programming contexts

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

biological and financial terms

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

 autorytatywna

-0.59

########.

-0.53

/**

-0.51

 AssemblyTitle

-0.50

SafeMath

-0.48

Rüyada

-0.48

PerformLayout

-0.47

 informée

-0.47

AsUp

-0.47

ंदीखरीदारी

-0.47

POSITIVE LOGITS

 rootReducer

0.38

 oreilles

0.36

dok

0.35

 Mitarbeit

0.33

stination

0.33

needs

0.32

manjaro

0.32

 sécher

0.32

Receiver

0.32

 vastaan

0.32

Activations Density 0.120%

specific names and identifiers related to structured data or programming contexts

biological and financial terms

No Comments

No Known Activations