INDEX

Explanations

references to images and concepts in technical or analytical contexts

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.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

EditorBrowsable

-0.52

 cref

-0.49

arraycopy

-0.49

 Bekannt

-0.48

brides

-0.48

atterson

-0.45

Бахар

-0.45

complexContent

-0.43

רת

-0.43

OutputPath

-0.42

POSITIVE LOGITS

 Underlying

0.95

underlying

0.88

 underlying

0.87

Underlying

0.86

 underlies

0.82

 underlie

0.81

 bakom

0.77

 originals

0.69

backing

0.67

 dietro

0.67

Activations Density 0.514%

references to images and concepts in technical or analytical contexts

No Comments

No Known Activations