INDEX

Explanations

parts of a document marked as comments or annotations

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

 betweenstory

-0.66

OGND

-0.54

 للاسماء

-0.53

AutoField

-0.51

bootstrapcdn

-0.50

hoeddwyd

-0.48

Autoritní

-0.48

 endforeach

-0.47

TestingModule

-0.44

 astore

-0.44

POSITIVE LOGITS

請繼續往下閱讀

0.59

 nakalista

0.56

 snippetHide

0.48

 HasFactory

0.39

astify

0.37

Viitteet

0.36

angering

0.36

setVerticalGroup

0.36

Erstellt

0.34

更好的

0.33

Activations Density 0.001%

parts of a document marked as comments or annotations

No Comments

No Known Activations