INDEX

Explanations

terms and phrases related to research, analysis, and data collection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 cons

-0.43

########.

-0.42

冑

-0.38

plaid

-0.36

 verte

-0.35

世間

-0.35

 inver

-0.35

ins

-0.35

 wort

-0.35

bed

-0.35

POSITIVE LOGITS

 jadx

0.58

 AssemblyCompany

0.48

 JADX

0.47

makeConstraints

0.47

 relâche

0.45

addGap

0.45

BeginContext

0.44

Rüyada

0.44

imageshack

0.42

 addSubview

0.42

Activations Density 0.680%

terms and phrases related to research, analysis, and data collection

No Comments

No Known Activations