INDEX

Explanations

boolean values and their representations in code

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

boolean true or yes

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

 Herz

-0.38

揭

-0.37

iba

-0.37

artiges

-0.36

 Researchers

-0.36

pernicus

-0.36

TERNS

-0.36

 halt

-0.36

hal

-0.35

 kaldı

-0.35

POSITIVE LOGITS

esModule

0.69

 defaultstate

0.59

 kasarigan

0.56

 Roskov

0.54

 autorytatywna

0.54

 AssemblyTitle

0.52

 nakalista

0.52

devamını

0.52

WriteBarrier

0.50

/**

0.50

Activations Density 0.002%

boolean values and their representations in code

boolean true or yes

No Comments

No Known Activations