INDEX

Explanations

references to conspiracies and collusion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 debit

-0.47

참고

-0.45

 catalog

-0.40

 exigences

-0.38

AppCompat

-0.38

 position

-0.37

();}

-0.36

 Veteran

-0.35

ijnlijk

-0.35

 departmental

-0.35

POSITIVE LOGITS

 conspiracy

0.76

conspiracy

0.70

 conspired

0.68

 conspira

0.67

 Conspiracy

0.62

 collusion

0.59

WebElementEntity

0.59

 conspir

0.58

MLLoader

0.56

 '\\;'

0.54

Activations Density 1.012%

references to conspiracies and collusion

No Comments

No Known Activations