INDEX

Explanations

questions that probe the validity or feasibility of a situation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 ویکی‌پدی

-0.65

دانشنامهٔ

-0.58

!*\

-0.58

 queſta

-0.58

principalColumn

-0.58

MessageTagHelper

-0.57

EDEFAULT

-0.57

 kasarigan

-0.57

irited

-0.57

IndentedString

-0.57

POSITIVE LOGITS

 whether

0.68

 是否

0.58

是否

0.56

 apakah

0.56

whether

0.56

 Apakah

0.55

 Whether

0.54

Whether

0.54

Apakah

0.53

是不是

0.52

Activations Density 0.370%

questions that probe the validity or feasibility of a situation

No Comments

No Known Activations