INDEX

Explanations

code import statements and directives from various programming languages

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

#include directives

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

AutoField

-0.68

GEBURTSDATUM

-0.66

AnchorStyles

-0.63

 <>",

-0.63

رشف

-0.62

-0.59

новниш

-0.58

posedge

-0.57

 OMITBAD

-0.56

MLLoader

-0.56

POSITIVE LOGITS

 fidé

0.39

 harita

0.39

 czeka

0.37

 magique

0.37

 sécur

0.36

 mourir

0.35

 śpiew

0.35

 seuls

0.33

 együtt

0.33

 seules

0.33

Activations Density 0.006%

code import statements and directives from various programming languages

#include directives

No Comments

No Known Activations