INDEX

Explanations

references to fakeness or deception, particularly in the context of news and representations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 primaire

-0.82

tamment

-0.72

 Audiodateien

-0.70

 disparu

-0.67

 lær

-0.67

 commissioners

-0.65

 behö

-0.65

CardHeader

-0.65

grà

-0.64

녔

-0.64

POSITIVE LOGITS

 fake

1.29

 Fake

1.20

Fake

1.11

fake

1.06

 pretend

0.98

 fakes

0.98

 Faux

0.94

Pret

0.93

 faux

0.93

Faux

0.90

Activations Density 0.333%

references to fakeness or deception, particularly in the context of news and representations

No Comments

No Known Activations