INDEX

Explanations

specific patterns or symbols indicating icons or buttons related to user actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

setVerticalGroup

-0.71

="@+

-0.66

 <<<<<<<<<<<<<<

-0.63

 autorytatywna

-0.63

 CreateTagHelper

-0.59

protoimpl

-0.57

 propOrder

-0.56

 GenerationType

-0.56

\{\\

-0.56

isible

-0.55

POSITIVE LOGITS

fa

0.56

fw

0.48

fa

0.42

 square

0.41

 magic

0.39

cog

0.37

Rog

0.37

 gears

0.37

 gear

0.37

users

0.36

Activations Density 0.099%

specific patterns or symbols indicating icons or buttons related to user actions

No Comments

No Known Activations