INDEX

Explanations

tokens that indicate page structure or formatting elements

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

GEBURTSDATUM

-1.57

 propOrder

-1.44

StoryboardSegue

-1.39

 Италијани

-1.28

ConstraintMaker

-1.28

IntoConstraints

-1.22

 виправивши

-1.18

Хьажоргаш

-1.16

AndEndTag

-1.14

modelBuilder

-1.13

POSITIVE LOGITS

\{\\

0.65

0.51

’

0.50

0.48

et

0.44

<i>

0.40

“

0.40

<eos>

0.40

 fact

0.39

Activations Density 0.021%

tokens that indicate page structure or formatting elements

No Comments

No Known Activations