INDEX

Explanations

metadata and categorizations of documents or data entries

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

rs sequence identifiers

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

tagHelperRunner

-0.83

ſelf

-0.78

ſelves

-0.70

 purpoſe

-0.66

expandindo

-0.59

 ſtate

-0.58

 AssemblyTitle

-0.58

AsUp

-0.56

 kaarangay

-0.56



-0.55

POSITIVE LOGITS

高达

0.42

 recently

0.42

bitField

0.41

 recientemente

0.41

 coroa

0.40

 older

0.40

 recentemente

0.39

recently

0.38

뀜

0.38

 cases

0.38

Activations Density 0.336%

metadata and categorizations of documents or data entries

rs sequence identifiers

No Comments

No Known Activations