INDEX

Explanations

numerical values and identifiers, likely related to error or status codes

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_131k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.33.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

 queſta

-0.81

IsContent

-0.75

 autorytatywna

-0.66

 otomatig

-0.66

IsMutable

-0.65

 виправивши

-0.64

 ujednoznacz

-0.63

 Majefty

-0.63

PreferredItem

-0.62

RTLI

-0.62

POSITIVE LOGITS

0.46

0.40

„

0.36

 chung

0.34

nut

0.32

kod

0.32

0.30

â

0.30

"/

0.30

@"

0.29

Activations Density 0.004%

numerical values and identifiers, likely related to error or status codes

No Comments

No Known Activations