INDEX

Explanations

tokens that indicate structural or programming elements in a document

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

áním

-0.51

 berço

-0.49

letus

-0.48

 alle

-0.47

kování

-0.47

 fièvre

-0.46

 Vorschriften

-0.46

bairro

-0.45

transport

-0.45

 amitié

-0.45

POSITIVE LOGITS

">//

0.95

SourceChecksum

0.87

 propOrder

0.80

 CreateTagHelper

0.79

TagMode

0.79

)");

0.78

 الرياضيه

0.76

 Roskov

0.71

*/;

0.71

migrationBuilder

0.69

Activations Density 0.363%

tokens that indicate structural or programming elements in a document

No Comments

No Known Activations