INDEX

Explanations

references to entities and relationships involving "to" and "for."

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

Japanese sentences containing particle "に"

np_acts-logits-general · gemini-2.0-flash

involved in or related to

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

行った

-0.73

行

-0.70

ある

-0.65

м

-0.61

的行

-0.61

行って

-0.60

おもしろ

-0.60

見る

-0.59

 élimin

-0.58

ly

-0.58

POSITIVE LOGITS

RenderAtEndOf

0.91

 esternos

0.75

 disambiguazione

0.70

 дописавши

0.67

]")]

0.66

DeleteBehavior

0.63

IBAction

0.59

 HasFactory

0.59

principalColumn

0.58

帖最后由

0.57

Activations Density 0.022%

references to entities and relationships involving "to" and "for."

Japanese sentences containing particle "に"

involved in or related to

No Comments

No Known Activations