INDEX

Explanations

terms related to speed and efficiency

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 enjoying

-0.61

requestCode

-0.58

addContainerGap

-0.55

 AppBar

-0.55

ArgsConstructor

-0.54

 rating

-0.53

 iprot

-0.53

 enjoy

-0.53

 enjoyed

-0.53

Aut

-0.52

POSITIVE LOGITS

Speed

1.22

speed

1.21

 Speed

1.13

SPEED

1.03

 induction

0.94

 speed

0.94

 SPEED

0.91

 Induction

0.82

 induc

0.81

Induction

0.80

Activations Density 0.158%

terms related to speed and efficiency

No Comments

No Known Activations