INDEX

Explanations

code related to file handling and imports in programming

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

archaic spellings using long s

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

charCode

-0.36

 Lake

-0.36

ConstraintMaker

-0.35

 Engine

-0.35

 Drink

-0.34

 الحره

-0.34

 Eric

-0.34

 Joshua

-0.34

Anak

-0.34

 Grave

-0.33

POSITIVE LOGITS

 InputDecoration

0.58

SequentialGroup

0.57

\{\\

0.52

 itſelf

0.50

ſelves

0.50

ſelf

0.49

 myſelf

0.49

Viited

0.49

forChild

0.49

 surla

0.48

Activations Density 0.002%

code related to file handling and imports in programming

archaic spellings using long s

No Comments

No Known Activations