INDEX

Explanations

statements challenging the validity of popular claims and highlighting hypocrisy in political discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/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.27.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

 QtCore

-0.45

 informée

-0.40

AnimationsModule

-0.39

 TextAlign

-0.38

eburger

-0.36

LEncoder

-0.36

HandlerContext

-0.34

fender

-0.34

 betweenstory

-0.34

splitContainer

-0.34

POSITIVE LOGITS

 bogus

0.79

 phony

0.76

 fraudulent

0.74

 falla

0.71

 laughable

0.70

 fake

0.69

 hoax

0.68

 sham

0.68

 ridiculous

0.68

 absurdity

0.67

Activations Density 0.577%

statements challenging the validity of popular claims and highlighting hypocrisy in political discourse

No Comments

No Known Activations