INDEX

Explanations

terms related to binary representations and data types, specifically in the context of programming and computer science

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

Integer or bit manipulation

np_acts-logits-general · gemini-2.0-flash

bits and integer types

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Konz

-0.45

wab

-0.42

неопр

-0.42

Viitteet

-0.42

 אס

-0.39

 photos

-0.39

 LUMP

-0.39

dtd

-0.39

Haj

-0.39

proba

-0.38

POSITIVE LOGITS

bit

1.03

bit

0.98

Bit

0.84

Bit

0.83

 intptr

0.80

 bits

0.79

 Overflow

0.79

 overflow

0.78

 Bits

0.77

 signo

0.74

Activations Density 0.820%

terms related to binary representations and data types, specifically in the context of programming and computer science

Integer or bit manipulation

bits and integer types

No Comments

No Known Activations