INDEX

Explanations

statements about conditions or states of being

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

"Is" followed by "about," "always," or "key"

np_acts-logits-general · gemini-2.0-flash

all about

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 ivelany

-0.73

ніципалі

-0.66

IndentedString

-0.65

tvguidetime

-0.64

 Réponses

-0.64

 betweenstory

-0.63

RegressionTest

-0.62

 nakalista

-0.58

queryInterface

-0.58

 Presumably

-0.57

POSITIVE LOGITS

 about

1.60

about

1.22

About

1.07

 ABOUT

1.07

 tentang

1.07

 About

1.03

ABOUT

0.90

关于

0.81

all

0.78

 predicated

0.77

Activations Density 0.256%

statements about conditions or states of being

"Is" followed by "about," "always," or "key"

all about

No Comments

No Known Activations