INDEX

Explanations

references to slices in imaging or data processing contexts

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 juuri

-0.71

 synagogue

-0.68

 Synagogue

-0.68

 pseudos

-0.64

 Fascism

-0.63

 Plume

-0.61

 comunale

-0.60

 lugs

-0.60

 hanged

-0.59

ANTLR

-0.59

POSITIVE LOGITS

ream

0.69

 Creator

0.68

Slice

0.66

slice

0.65

 creator

0.64

 Slice

0.62

createStatement

0.61

 beginnetje

0.61

 Creators

0.59

oredCriteria

0.57

Activations Density 0.089%

references to slices in imaging or data processing contexts

No Comments

No Known Activations