INDEX

Explanations

mathematical expressions and symbols related to networks and their structures

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

edges between nodes

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

NTIS

-0.32

허

-0.32

 answer

-0.32

AndroidJUnit

-0.32

 définiti

-0.31

uterie

-0.29

Geplaatst

-0.29

cabulary

-0.29

GetEnumerator

-0.28

nextNode

-0.27

POSITIVE LOGITS

HasIndex

0.65

.*")]

0.65

wixt

0.60

 BETWEEN

0.60

enumii

0.58

BETWEEN

0.58

 between

0.58

fjspx

0.56

 zwischen

0.54

owohl

0.52

Activations Density 0.128%

mathematical expressions and symbols related to networks and their structures

edges between nodes

No Comments

No Known Activations