INDEX

Explanations

references to hierarchical or nested structures and their attributes

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

entity id path genre parent

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

MessageTagHelper

-0.64

chartInstance

-0.49

økt

-0.47

 numberOfRows

-0.46

consulté

-0.45

 HasFactory

-0.44

LikeLiked

-0.44

Hozzáférés

-0.44

 Infórmanos

-0.42

__))

-0.42

POSITIVE LOGITS

engesch

0.40

 linkovi

0.38

ideae

0.37

ünstig

0.37

 transfieras

0.37

 kanyang

0.35

ginate

0.34

esign

0.34

 lenker

0.34

machus

0.34

Activations Density 0.030%

references to hierarchical or nested structures and their attributes

entity id path genre parent

No Comments

No Known Activations