INDEX

Explanations

technical and scientific terminology in various contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

enumii

-0.56

 pleaſure

-0.49

 NSCoder

-0.49

 Perſ

-0.45

 betweenstory

-0.44

MessageWindow

-0.43

日閲覧

-0.43

ième

-0.42

 zamiast

-0.42

terday

-0.42

POSITIVE LOGITS

ChildScrollView

0.47

Aholisi

0.46

 ComVisible

0.39

mstyle

0.38

TintMode

0.34

ียญ

0.34

bud

0.32

addrs

0.32

Raw

0.31

 bụi

0.31

Activations Density 0.755%

technical and scientific terminology in various contexts

No Comments

No Known Activations