INDEX

Explanations

references to URLs and web-related actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Reſ

-0.85

 ſche

-0.78

 Majefty

-0.75

 Inſ

-0.74

 houſe

-0.74

 raiſ

-0.72

 ſtate

-0.72

 pleaſure

-0.72

 itſelf

-0.71

 ſever

-0.71

POSITIVE LOGITS

 Google

0.85

www

0.85

 Amazon

0.74

Google

0.71

 google

0.70

 sites

0.69

 eBay

0.68

www

0.66

 sitios

0.64

 YouTube

0.62

Activations Density 0.381%

references to URLs and web-related actions

No Comments

No Known Activations