INDEX

Explanations

phrases that highlight statistical rankings or high levels of certain measurements or conditions

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

highest rates or percentages

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

TKey

-0.46

ędziesz

-0.44

tronics

-0.43

teiro

-0.42

IndentedString

-0.42

estro

-0.41

SD

-0.41

gridy

-0.41

poti

-0.40

 HttpNotFound

-0.40

POSITIVE LOGITS

SourceChecksum

0.53

oneofs

0.47

ValueStyle

0.45

fitriones

0.43

0.41

 ujednoznacz

0.40

 betweenstory

0.40

ResumeLayout

0.40

StoryboardSegue

0.39

 allAfrica

0.38

Activations Density 0.011%

phrases that highlight statistical rankings or high levels of certain measurements or conditions

highest rates or percentages

No Comments

No Known Activations