INDEX

Explanations

segments of text related to legal or formal examination processes

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

direct examination

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

expandindo

-0.70

tvguidetime

-0.62

complexContent

-0.59

 AssemblyCulture

-0.58

IUrlHelper

-0.55

+#+#

-0.49

 виправивши

-0.48

 дописавши

-0.48

 invokingState

-0.48

Personensuche

-0.48

POSITIVE LOGITS

MLLoader

0.43

 doInBackground

0.41

wapV

0.40

RTLU

0.40

enderror

0.38

LabelTagHelper

0.37

正

0.37

 ammat

0.37

 AssemblyTitle

0.37

emptory

0.37

Activations Density 0.073%

segments of text related to legal or formal examination processes

direct examination

No Comments

No Known Activations