INDEX

Explanations

references to methods and details related to scientific procedures

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

for details

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

tagHelperRunner

-0.64

MessageTagHelper

-0.61

 חיצוניים

-0.59

 resourceCulture

-0.56

DockStyle

-0.55

parsedMessage

-0.54

 protoimpl

-0.50

 queſta

-0.50

 Budget

-0.50

باردا

-0.49

POSITIVE LOGITS

 specifics

0.39

 intptr

0.39

 details

0.39

 noqa

0.38

 restrictions

0.37

 contains

0.37

 Auswirkungen

0.36

 reasons

0.35

 requirements

0.35

 tanggung

0.35

Activations Density 0.028%

references to methods and details related to scientific procedures

for details

No Comments

No Known Activations