INDEX

Explanations

references to lists, codes, and specific programming or technical terms

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aarrggbb

-0.84

 nakalista

-0.71

 OMIT

-0.68

 betweenstory

-0.63

UpInside

-0.62

AsUp

-0.59

mybatisplus

-0.58

GraphicsUnit

-0.57

TargetException

-0.55

 iprot

-0.55

POSITIVE LOGITS

SEGUIR

0.56

 itſelf

0.56

 Jefus

0.53

 Houſe

0.51

 Chriſt

0.49

 aprovecha

0.49

ècie

0.49

 decoracion

0.49

mlung

0.49

 ſay

0.49

Activations Density 0.005%

references to lists, codes, and specific programming or technical terms

No Comments

No Known Activations