INDEX

Explanations

terms related to data types and media types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 protoimpl

-0.83

 Theſe

-0.72

 Paglinawan

-0.68

 Cæsar

-0.66

تقاوى

-0.65

 venons

-0.65

Alike

-0.63

rawDesc

-0.62

bkz

-0.62

CppMethod

-0.61

POSITIVE LOGITS

 Hours

0.64

typedef

0.57

types

0.57

 minutes

0.56

ilever

0.56

 types

0.53

fix

0.53

Issue

0.53

ButterKnife

0.53

 Officers

0.52

Activations Density 0.232%

terms related to data types and media types

No Comments

No Known Activations