INDEX

Explanations

phrases related to statistical measures and data ranges

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

near "max", "min", "mean", or "range"

np_acts-logits-general · gemini-2.0-flash

min, max, and range of values

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

eren

-0.45

cés

-0.37

進

-0.36

bilità

-0.35

van

-0.34

 Fenton

-0.34

 franco

-0.34

ere

-0.34

子

-0.33



-0.32

POSITIVE LOGITS

 maximum

2.05

max

1.98

max

1.83

maximum

1.82

 minimum

1.75

 Maximum

1.70

 highest

1.68

 maximal

1.65

 lowest

1.64

Maximum

1.63

Activations Density 0.555%

phrases related to statistical measures and data ranges

near "max", "min", "mean", or "range"

min, max, and range of values

No Comments

No Known Activations