INDEX

Explanations

patterns or structure in scientific or technical data

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

Numbers within parentheses

np_acts-logits-general · gemini-2.0-flash

numeric ranges and averages

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

 posi

-0.53

 propOrder

-0.49

 hundreds

-0.48

 thousands

-0.48

 Posi

-0.46

rande

-0.46

hibernate

-0.44

xffff

-0.43

cava

-0.43

 Stad

-0.42

POSITIVE LOGITS

Average

0.76

 average

0.70

EndContext

0.69

verages

0.68

enderror

0.68

 averages

0.67

average

0.67

parsedMessage

0.66

 Average

0.65

})*/

0.64

Activations Density 0.404%

patterns or structure in scientific or technical data

Numbers within parentheses

numeric ranges and averages

No Comments

No Known Activations