INDEX

Explanations

code elements related to variable definitions and comparisons within programming or scripting contexts

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

new code and math

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

 OFDb

-0.73

AutoScaleMode

-0.73

 autorytatywna

-0.71

 stockbild

-0.64

 Савезне

-0.63

AccessorTable

-0.63

sidemargin

-0.62

RegressionTest

-0.59

 cherchés

-0.58

 &___

-0.58

POSITIVE LOGITS

 ब्रेकडाउन

0.45

Manbalar

0.36

Földrajz

0.27

new

0.27

cipais

0.26

 passing

0.26

 everywhere

0.26

 passed

0.25

 nuevos

0.24

 emotionally

0.24

Activations Density 0.033%

code elements related to variable definitions and comparisons within programming or scripting contexts

new code and math

No Comments

No Known Activations