INDEX

Explanations

patterns or mentions of numerical data and figures

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

httphttps

-0.91

+#+#

-0.86

 CreateTagHelper

-0.78

findpost

-0.77

bootstrapcdn

-0.77

 للاسماء

-0.74

tagHelperRunner

-0.73

 surla

-0.70

 '\\;'

-0.69

 Infórmanos

-0.69

POSITIVE LOGITS

AutoField

0.32

0.27

0.26

kembangan

0.26

0.25

0.24

CP

0.24

Activations Density 0.021%

patterns or mentions of numerical data and figures

No Comments

No Known Activations