INDEX

Explanations

references to scientific research and observational studies involving various entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

遍

-0.49

/*---

-0.49

gep

-0.48

jgl

-0.46

enol

-0.45

 business

-0.45

Working

-0.44

WriteLiteral

-0.43

ת

-0.43

Tas

-0.43

POSITIVE LOGITS

OGND

0.80

principalTable

0.76

IUrlHelper

0.76

 <=",

0.73

 للاسماء

0.71

 تضيفلها

0.70

AntiForgeryToken

0.68

 незавершена

0.66

Diwedd

0.66

 AssemblyCompany

0.65

Activations Density 0.490%

references to scientific research and observational studies involving various entities

No Comments

No Known Activations