INDEX

Explanations

words expressing degrees of comparison or changes in magnitude

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

After the word "greater"

np_acts-logits-general · gemini-2.0-flash

greater degree or scale

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

ocup

-0.54

HandlerContext

-0.47

 Dase

-0.46

DOCTYPE

-0.45

万

-0.44

ПП

-0.44

安い

-0.43

 newer

-0.43

abro

-0.42

 crappy

-0.42

POSITIVE LOGITS

 amount

1.08

amount

0.91

 amounts

0.85

 degree

0.83

 الحره

0.81

 number

0.80

ngdoc

0.79

 complexity

0.75

amounts

0.75

 quantity

0.73

Activations Density 0.691%

words expressing degrees of comparison or changes in magnitude

After the word "greater"

greater degree or scale

No Comments

No Known Activations