INDEX

Explanations

phrases indicating examples or instances

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

example and instance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

AxisAlignment

-0.58

writerow

-0.47

 licensed

-0.41

 Reality

-0.41

burgs

-0.38

 professional

-0.37

Reality

-0.36

 licensees

-0.36

olymer

-0.35

 Fuzzy

-0.35

POSITIVE LOGITS

WireFormatLite

0.51

Diwedd

0.50

 >=",

0.49

protoimpl

0.47

 AssemblyCulture

0.47

########.

0.43

UseVisualStyle

0.43

intenant

0.43

 كومونز

0.42

󠁢

0.42

Activations Density 0.002%

phrases indicating examples or instances

example and instance

No Comments

No Known Activations