INDEX

Explanations

key terms related to official resolutions or actions

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

official titles and grants

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

XmlAccessorType

-0.55

bootstrapcdn

-0.52

ArrowToggle

-0.50

çons

-0.47

SuppressLint

-0.46

boolean

-0.45

 nhật

-0.44

bargain

-0.44

nucle

-0.42

awtextra

-0.42

POSITIVE LOGITS

 Италијани

0.40

 ControllerBase

0.38

 Gebä

0.36

 enfermed

0.34

rungsseite

0.34

 namanya

0.34

 wystaw

0.34

 reconoce

0.33

 recognizing

0.33

 recognition

0.33

Activations Density 0.015%

key terms related to official resolutions or actions

official titles and grants

No Comments

No Known Activations