INDEX

Explanations

phrases indicating recognition or acknowledgment of achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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



-0.83

 Chwiliwch

-0.81

Xna

-0.76

UserScript

-0.73

enderror

-0.70

CUC

-0.68

 الرياضيه

-0.68

ⓧ

-0.67

Bux

-0.67

 Jove

-0.67

POSITIVE LOGITS

égard

0.51

</u>

0.43

 versus

0.43

0.42

Silva

0.42

vs

0.41

Kurz

0.41

lingua

0.41

did

0.40

させていただきました

0.39

Activations Density 0.271%

phrases indicating recognition or acknowledgment of achievements

No Comments

No Known Activations