INDEX

Explanations

terms related to coding and software functionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

CloseOperation

-0.66

 pleaſure

-0.60

 sternum

-0.59

 elongation

-0.59

 adjournment

-0.58

 Hermans

-0.58

migo

-0.58

HideFlags

-0.57

 spatula

-0.57

 ſeveral

-0.57

POSITIVE LOGITS

or

0.62

 were

0.57

you

0.54

 quelconque

0.52

is

0.52

 saites

0.52

 ever

0.51

дъ

0.50

 fails

0.48

 hadn

0.48

Activations Density 0.548%

terms related to coding and software functionality

No Comments

No Known Activations