INDEX

Explanations

references to licensing terms and categories related to academic articles

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

RegressionTest

-0.98

 Савезне

-0.85

 CreateTagHelper

-0.83

 jsPsych

-0.81

 kasarigan

-0.81

 &___

-0.81

UnusedPrivate

-0.81

NameInMap

-0.81

mybatisplus

-0.79

oredCriteria

-0.79

POSITIVE LOGITS

 Commons

1.11

Wikimedia

1.03

Commons

0.86

 Wikimedia

0.84

 commons

0.84

wikimedia

0.79

ExtendWith

0.62

commons

0.61

imedia

0.57

 Images

0.57

Activations Density 0.053%

references to licensing terms and categories related to academic articles

No Comments

No Known Activations