INDEX

Explanations

transitions or clarifications concerning claims and evidence in discussions or arguments

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

AppMethodBeat

-0.38

Vegeu

-0.31

ppuden

-0.31

qJ

-0.29

krieg

-0.29

 Privacidade

-0.28

ActiveRecord

-0.28

BeginContext

-0.27

 nôtre

-0.27

POSITIVE LOGITS

SequentialGroup

0.66

 Савезне

0.59

WebElementEntity

0.58

<unused79>

0.57

<unused14>

0.57

[@BOS@]

0.57

<unused41>

0.57

<unused51>

0.57

<unused17>

0.57

<unused8>

0.57

Activations Density 1.285%

transitions or clarifications concerning claims and evidence in discussions or arguments

No Comments

No Known Activations