INDEX

Explanations

unusual formatting or patterns in text

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

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

хьтан

-0.87

Autoritní

-0.70

 nahilalakip

-0.68

webElementXpaths

-0.65

AnimationsModule

-0.63

 autorytatywna

-0.63

 يتيمه

-0.62

featureID

-0.59

Atsauces

-0.57

Хьажоргаш

-0.56

POSITIVE LOGITS

vol

0.45

 Fixation

0.42

Gae

0.38

 groupBox

0.37

 assembler

0.36

 Manifest

0.36

 Clause

0.36

decoration

0.35

 Message

0.35

acar

0.35

Activations Density 0.170%

unusual formatting or patterns in text

No Comments

No Known Activations