INDEX

Explanations

the concept of "intrinsic" attributes or properties

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

 intrinsic

-1.94

 Intrinsic

-1.77

intrinsic

-1.73

Intrinsic

-1.65

 intrins

-1.44

 extrinsic

-1.28

 intrinsically

-1.13

insic

-0.95

 Intr

-0.83

Intr

-0.80

POSITIVE LOGITS

eclamp

0.40

CrossRef

0.40

 Schlä

0.39

 обору

0.39

itinéraire

0.38

zelt

0.38

 Shetty

0.37

isNullOrEmpty

0.37

 PROCESSES

0.37

FACE

0.37

Activations Density 0.003%

the concept of "intrinsic" attributes or properties

No Comments

No Known Activations