INDEX

Explanations

references to secrecy and mystery, particularly regarding historical or legendary groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 <<<<<<<<<<<<<<

-0.73

AddTagHelper

-0.64

 createSprite

-0.63

Посилання

-0.61

enumii

-0.60

mybatisplus

-0.59

Aiheesta

-0.59

TypedDataSet

-0.58

NameInMap

-0.58

出版年

-0.57

POSITIVE LOGITS

 mysterious

0.95

 mystery

0.91

 mysteries

0.80

mystery

0.80

 secretive

0.78

 secrecy

0.77

 secret

0.75

 shrouded

0.74

 unknown

0.74

Mystery

0.73

Activations Density 0.271%

references to secrecy and mystery, particularly regarding historical or legendary groups

No Comments

No Known Activations