INDEX

Explanations

terms related to causation and influence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 ब्रेकडाउन

-0.72

WebElementEntity

-0.59

ValueStyle

-0.59

 DBNull

-0.51

 kaarangay

-0.50

migrationBuilder

-0.50

 referrerpolicy

-0.48

 otomatig

-0.47

 surla

-0.46

 aload

-0.46

POSITIVE LOGITS

AutoField

0.60

ChildScrollView

0.60

SharedDtor

0.47

Бахар

0.44

ніципа

0.44

 NSCoder

0.43

sidemargin

0.40

Lähde

0.35

 triggered

0.35

 meningkatkan

0.34

Activations Density 1.684%

terms related to causation and influence

No Comments

No Known Activations