INDEX

Explanations

phrases and words related to the generation of various outputs or objects

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

generate content or files

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

XmlAccessType

-0.71

 OkHttpClient

-0.54

 ujednoznacz

-0.53

näm

-0.53



-0.52

URLConnection

-0.50

égard

-0.49

 étro

-0.49

AnchorStyles

-0.48

aspectj

-0.48

POSITIVE LOGITS

generate

0.81

 generate

0.76

 generation

0.68

Generate

0.67

 generated

0.64

 creation

0.63

 Generate

0.62

 generator

0.62

生成

0.62

 generating

0.58

Activations Density 0.022%

phrases and words related to the generation of various outputs or objects

generate content or files

No Comments

No Known Activations