INDEX

Explanations

words related to secrecy and hidden knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 AssemblyCulture

-0.74

BeginInit

-0.74

TagMode

-0.73

nologue

-0.70

 Frazier

-0.67

makeText

-0.66

 endwhile

-0.66

unhofer

-0.65

 ujednoznacz

-0.63

 BoxFit

-0.62

POSITIVE LOGITS

 secret

0.90

secret

0.88

Secret

0.85

 secreta

0.79

 Secret

0.79

delegate

0.76

Delegate

0.76

 lease

0.76

 timely

0.74

 disclosure

0.73

Activations Density 0.064%

words related to secrecy and hidden knowledge

No Comments

No Known Activations