INDEX

Explanations

references to safety and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

complexContent

-1.01

 TableColumn

-0.81

 Hooper

-0.80

prägt

-0.75

AddTagHelper

-0.75

 McCulloch

-0.74

RegressionTest

-0.72

IndentedString

-0.70

JMenu

-0.69

 pró

-0.69

POSITIVE LOGITS

 safety

1.32

SAFE

1.26

 safe

1.26

 Safety

1.25

SAFETY

1.25

afety

1.24

 Safe

1.24

Safety

1.21

 SAFETY

1.16

 safer

1.15

Activations Density 0.063%

references to safety and related concepts

No Comments

No Known Activations