INDEX

Explanations

comparisons and rankings among entities or concepts

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

topping or outperforming others

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

 déri

-0.38

ideen

-0.35

 îna

-0.35

dă

-0.34

 isolato

-0.34

 merav

-0.34

izzato

-0.34

 titolata

-0.33

 miejsco

-0.33

 asciug

-0.33

POSITIVE LOGITS

elemField

0.63

EndContext

0.57

MemoryWarning

0.56

فایل‌لار

0.52

ThemeOverlay

0.51

 surla

0.50

 AssemblyTitle

0.46

ớ

0.45

 fart

0.44

fjspx

0.44

Activations Density 0.116%

comparisons and rankings among entities or concepts

topping or outperforming others

No Comments

No Known Activations