INDEX

Explanations

occurrences of the word "to" in various contexts

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

infinitive verbs

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

 BrowserModule

-0.46

sizePolicy

-0.46

 enfans

-0.45

 Consideration

-0.44

 gyhoeddwyd

-0.42

printStackTrace

-0.42

DialogContent

-0.41

 capitolo

-0.40

extrême

-0.39

 manžel

-0.39

POSITIVE LOGITS

EndInit

0.58

 help

0.55

を使って

0.49

 smugglers

0.47

 помогают

0.45

さで

0.45

DockStyle

0.43

 تضيفلها

0.43

yyj

0.43

 create

0.43

Activations Density 0.043%

occurrences of the word "to" in various contexts

infinitive verbs

No Comments

No Known Activations