INDEX

Explanations

scientific classifications and taxonomic information

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

 purpoſe

-0.51

Попис

-0.47

GEBURTS

-0.47



-0.45

Datuak

-0.45

windowFixed

-0.43

Chham

-0.41

 MediaType

-0.41

 lode

-0.41

IVEREF

-0.41

POSITIVE LOGITS

 tangan

0.50

 Referenzen

0.46

 Água

0.46

tilha

0.45

 düny

0.45

 betrek

0.45

 noiva

0.43

 paisagem

0.43

 tarto

0.43

 inspiradoras

0.42

Activations Density 0.037%

scientific classifications and taxonomic information

No Comments

No Known Activations