INDEX

Explanations

phrases indicating comparisons or contrasts between different options or features

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

Comes before descriptions of severity or magnitude

np_acts-logits-general · gemini-2.0-flash

advanced, larger, higher

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Италијани

-0.61

SIMPLE

-0.61

 innocent

-0.60

 Utilizamos

-0.59

简单

-0.58

VersionUID

-0.57

 innocence

-0.56

 sederhana

-0.56

 простой

-0.56

Билгалдахарш

-0.55

POSITIVE LOGITS

 advanced

1.34

 larger

1.26

 bigger

1.25

 higher

1.24

advanced

1.20

 Advanced

1.13

Advanced

1.13

 Larger

1.10

 ADVANCED

1.09

 sophisticated

1.08

Activations Density 1.115%

phrases indicating comparisons or contrasts between different options or features

Comes before descriptions of severity or magnitude

advanced, larger, higher

No Comments

No Known Activations