INDEX

Explanations

numeric values and symbols, particularly focusing on negative values and their context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

AndEndTag

-0.72

KindOfClass

-0.57

 Cavity

-0.57

 Flach

-0.57

")->

-0.56

daction

-0.55

Aar

-0.54

Kaf

-0.54

IFORNIA

-0.53

tyimages

-0.53

POSITIVE LOGITS

 ostavi

0.68

InjectAttribute

0.62

expandindo

0.58

queryInterface

0.56

contentLoaded

0.55

Normdaten

0.54

 Roskov

0.53

 estimés

0.53

 متعلقه

0.51

kloped

0.51

Activations Density 0.021%

numeric values and symbols, particularly focusing on negative values and their context

No Comments

No Known Activations