INDEX

Explanations

negations and phrases that indicate the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

})()

-0.56

RenderAtEndOf

-0.55

 jopa

-0.55

dataclass

-0.54

styleType

-0.54

makeConstraints

-0.53

 recettes

-0.52

fjspx

-0.51

 triplicate

-0.51

今度は

-0.51

POSITIVE LOGITS

 directly

0.73

دانشنامهٔ

0.69

 formally

0.68

any

0.67

 noDo

0.66

 necessarily

0.64

 officially

0.59

 formal

0.59

 explicitly

0.58

 ProtoMessage

0.57

Activations Density 0.708%

negations and phrases that indicate the absence of something

No Comments

No Known Activations