INDEX

Explanations

elements related to observational studies and their results

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

out of, annoys, struck

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

ChildScrollView

-0.46

featureID

-0.45

 purpoſe

-0.44

 preſent

-0.44

 ſtand

-0.43

يميديا

-0.43

 Diſ

-0.42

Jäh

-0.41

 الحره

-0.41

RepeatedField

-0.41

POSITIVE LOGITS

ьаж

0.39

 Italijanski

0.38

 Wicidata

0.37

Autoritní

0.37

Geplaatst

0.37

 costo

0.36

 Stuff

0.36

 definition

0.35

 games

0.35

 experience

0.35

Activations Density 0.214%

elements related to observational studies and their results

out of, annoys, struck

No Comments

No Known Activations