INDEX

Explanations

elements related to programming and code structure

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

turn user

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

RegressionTest

-0.76

SourceChecksum

-0.68

writeFieldEnd

-0.62

 typelib

-0.62

rungsseite

-0.60

elemField

-0.59

 surla

-0.54

ⓧ

-0.53

المناصب

-0.52

 estekak

-0.51

POSITIVE LOGITS

 AttributeSet

1.11

AttributeSet

0.51

 urbaine

0.40

 procès

0.36

 rynku

0.35

 carteira

0.35

InteropServices

0.35

 tombé

0.35

kregen

0.34

 nahilalakip

0.34

Activations Density 0.001%

elements related to programming and code structure

turn user

No Comments

No Known Activations