INDEX

Explanations

programmatic elements related to object-oriented programming and memory management

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

Code or names followed by hyphens or underscores

np_acts-logits-general · gemini-2.0-flash

constructor-arg, small-caps, unsigned_long

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

########.

-0.70

 Wikispecies

-0.61

 RIPRODUZIONE

-0.57

 Савезне

-0.56

BagConstraints

-0.56

 Roskov

-0.56

 محفوظة

-0.55

spesies

-0.55

 soeur

-0.52

findpost

-0.52

POSITIVE LOGITS

0.95

0.80

0.64

amp

0.63

imageshack

0.63

InlineData

0.62

‌

0.61

・

0.58

 surla

0.53

plona

0.52

Activations Density 1.256%

programmatic elements related to object-oriented programming and memory management

Code or names followed by hyphens or underscores

constructor-arg, small-caps, unsigned_long

No Comments

No Known Activations