INDEX

Explanations

references to vague or unspecified items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Eres

-0.52

 GetEnumerator

-0.52

 Elles

-0.51

Revenir

-0.51

IndentedString

-0.50

 humeur

-0.50

тесь

-0.50

✨:

-0.48

ください

-0.47

 piatta

-0.47

POSITIVE LOGITS

migrationBuilder

0.69

 CreateTagHelper

0.63

 bezeichneter

0.55

ⓧ

0.55

 SwiftUI

0.53

zitter

0.52

牧

0.51

inose

0.51

ürlich

0.51

oflavin

0.51

Activations Density 0.139%

references to vague or unspecified items

No Comments

No Known Activations