INDEX

Explanations

quotations or direct speech in the text

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

AndEndTag

-0.67

Eſ

-0.62

SBATCH

-0.61

CDP

-0.59

InstanceState

-0.55

 itſelf

-0.50

 chaus

-0.49

Zuz

-0.49

 Sopho

-0.49

SPE

-0.48

POSITIVE LOGITS

 насељу

0.70

bootstrapcdn

0.61

ArrowToggle

0.61

Autoritní

0.59

FunctionFlags

0.57

migrationBuilder

0.56

smithy

0.56

 kasarigan

0.56

كويكب

0.55

 ddelweddau

0.55

Activations Density 0.105%

quotations or direct speech in the text

No Comments

No Known Activations