INDEX

Explanations

elements related to user registration and naming conventions in a system

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

technical terms and placeholders

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

rungsseite

-0.81

NameInMap

-0.75

Autoritní

-0.71

adpleegd

-0.66

XmlAccessorType

-0.63

 intStringLen

-0.61

 GenerationType

-0.60

nodoc

-0.60

 dAtA

-0.60

 propOrder

-0.60

POSITIVE LOGITS

 required

0.33

 felicità

0.33

 wcześ

0.33

 failure

0.32

 failed

0.31

 חי

0.30

 concentration

0.30

jsdelivr

0.30

 clearance

0.29

sweetalert

0.29

Activations Density 0.007%

elements related to user registration and naming conventions in a system

technical terms and placeholders

No Comments

No Known Activations