INDEX

Explanations

terms related to classification and categorization, particularly within academic and scientific contexts

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

levels, scales, and types

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

 comigo

-0.42

étoit

-0.39

guien

-0.39

 âmes

-0.38

งาม

-0.38

vœ

-0.38

 segredo

-0.38

 enfans

-0.37

ByKey

-0.37

ándola

-0.36

POSITIVE LOGITS

yntaxException

0.67

Statistiche

0.60

 AssemblyCompany

0.59

httphttps

0.59

AndEndTag

0.57

nthetic

0.56

postsleuth

0.55



0.55

الحياه

0.55

Portale

0.54

Activations Density 0.048%

terms related to classification and categorization, particularly within academic and scientific contexts

levels, scales, and types

No Comments

No Known Activations