INDEX

Explanations

phrases related to medical and organizational contexts

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

started out, originally

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

➯

-0.45

WebMethod

-0.44

 CURIAM

-0.44

basicConfig

-0.44

AddTagHelper

-0.43

AccessorTable

-0.43

 defaultstate

-0.42

曖昧さ回避

-0.42

اریخ

-0.42

 Roskov

-0.42

POSITIVE LOGITS

 ursprünglich

0.45

 proposons

0.44

Originally

0.42

 simple

0.42

 originally

0.41

 teething

0.41

 purely

0.40

ImageContext

0.40

 originalmente

0.39

СТО

0.39

Activations Density 0.032%

phrases related to medical and organizational contexts

started out, originally

No Comments

No Known Activations