INDEX

Explanations

instances of section or topic headings typically found in academic or technical writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

RenderAtEndOf

-0.88

Eſ

-0.86

Efq

-0.83

 Majefty

-0.79

 Monfieur

-0.79

 becauſe

-0.78

 Theſe

-0.76

 Anſ

-0.75

 незавершена

-0.74

 ſtate

-0.73

POSITIVE LOGITS

LabelTagHelper

0.57

 mendatang

0.54

IUrlHelper

0.48

or

0.47

ttino

0.47

میل

0.46

Jornal

0.45

if

0.45

Enders

0.44

Activations Density 0.049%

instances of section or topic headings typically found in academic or technical writing

No Comments

No Known Activations