INDEX

Explanations

words indicating medical or judicial conditions and assessments

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

ParseTree

-0.37

 yerine

-0.36

 başlay

-0.35

to

-0.35

ARROLLO

-0.34

 plă

-0.33

all

-0.33

POSITIVE LOGITS

 Meksiku

0.70

pecabe

0.66

ſehen

0.65

InjectAttribute

0.63

 beſte

0.63



0.62

 mauva

0.62

 Geiſt

0.60

addCriterion

0.60

 ſind

0.59

Activations Density 0.017%

words indicating medical or judicial conditions and assessments

No Comments

No Known Activations