INDEX

Explanations

articles and variations of the word "a" in text

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

describing notable events or outcomes

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

TableAdapter

-0.41

 corporations

-0.40

 institu

-0.37

layoutControl

-0.36

 culture

-0.35

 consumer

-0.35

 society

-0.35

 pleaſure

-0.35

consumer

-0.35

Detroit

-0.34

POSITIVE LOGITS

ConstraintMaker

0.64

 spectacular

0.59

 exhilarating

0.57

fromCharCode

0.56

rieks

0.56

 amazingly

0.55

 thrilling

0.55

 surprisingly

0.54

 zweimal

0.54

 remarkably

0.53

Activations Density 0.011%

articles and variations of the word "a" in text

describing notable events or outcomes

No Comments

No Known Activations