INDEX

Explanations

proper nouns related to companies and legal cases

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

named entities and following words

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

 мәкал

-0.77

StoryboardSegue

-0.73

UserScript

-0.70

GEBURTSDATUM

-0.69

postsleuth

-0.66

 AssemblyProduct

-0.65

 Chwiliwch

-0.62

RegressionTest

-0.62

tvguidetime

-0.61

-------------</

-0.61

POSITIVE LOGITS

ฯ

0.47

 Wikiseite

0.45

,:);

0.35

 mukaan

0.34

 antaranya

0.34

 său

0.34

RenderAtEndOf

0.34

 beliau

0.34

 ettei

0.33

 meille

0.33

Activations Density 0.045%

proper nouns related to companies and legal cases

named entities and following words

No Comments

No Known Activations