INDEX

Explanations

estimated quantities and ratios in a context involving care and management

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

Preceding "less" or indicating a smaller amount

np_acts-logits-general · gemini-2.0-flash

more or less

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

DockStyle

-0.83

 الحره

-0.74

 CreateTagHelper

-0.68

ModelSerializer

-0.66

 дописавши

-0.66

ConstraintMaker

-0.66

RegressionTest

-0.64

flink

-0.64

 ModelExpression

-0.62

POSITIVE LOGITS

 less

3.60

 Less

3.17

Less

3.01

 moins

2.70

 menos

2.65

 LESS

2.64

less

2.55

 fewer

2.36

 weniger

2.28

LESS

2.21

Activations Density 0.536%

estimated quantities and ratios in a context involving care and management

Preceding "less" or indicating a smaller amount

more or less

No Comments

No Known Activations