INDEX

Explanations

underscore characters used for formatting or emphasis

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

úrese

-0.67

=?";

-0.67

 himo

-0.66

)";

-0.66

énario

-0.65

 iprot

-0.64

{}",

-0.63

)";

-0.63

{}".

-0.63

 препратки

-0.63

POSITIVE LOGITS

 pleaſure

0.73

+"_

0.71

 cilia

0.71

 Elbe

0.69

DockStyle

0.69

 ſtate

0.69

0.67

 Pliocene

0.66

 laſt

0.66

 Suivez

0.66

Activations Density 0.004%

underscore characters used for formatting or emphasis

No Comments

No Known Activations