INDEX

Explanations

references to owls or owl-related content

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

owl

-1.77

Owl

-1.66

 owls

-1.47

Owl

-1.47

owl

-1.39

 Owls

-1.23

OWL

-1.22

OWL

-1.15

🦉

-0.73

Ow

-0.45

POSITIVE LOGITS

ArgsConstructor

0.54

PickerController

0.49

LogFactory

0.48

 Indes

0.46

Configurable

0.45

 ContentValues

0.44

daction

0.44

 StringTokenizer

0.44

CodeAttribute

0.43

 Cases

0.43

Activations Density 0.001%

references to owls or owl-related content

No Comments

No Known Activations