INDEX

Explanations

occurrences of the letter "F" and other specific procedural elements or indicators in a document

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

F followed by Par

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

 beginnetje

-0.67

 незавершена

-0.60

 kaynağından

-0.59

 дописавши

-0.59

المناصب

-0.56

 الحره

-0.54

 Wikimedijinoj

-0.52

 وتسجيلات

-0.52

Geplaatst

-0.50

principalColumn

-0.50

POSITIVE LOGITS

bibfield

0.52

addProduct

0.46

BibitemShut

0.45

Unspecified

0.43

bibinfo

0.41

cerne

0.40

 avoient

0.40

 poffible

0.39

setTotal

0.39

 UNIVERSAL

0.39

Activations Density 0.001%

occurrences of the letter "F" and other specific procedural elements or indicators in a document

F followed by Par

No Comments

No Known Activations