INDEX

Explanations

terms or phrases indicating classifications or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

 Grüsse

-0.56

UrlResolution

-0.53

 asupra

-0.52

latego

-0.52

 zufolge

-0.51

 mediodía

-0.50

principalTable

-0.47

 tartalomajánló

-0.47

 beschik

-0.47

 antemano

-0.46

POSITIVE LOGITS

 trip

0.63

###

0.58

 Alice

0.58

getSystemService

0.55

 pair

0.55

lot

0.54

 plot

0.54

GEBURTSDATUM

0.54

 species

0.52

 bash

0.52

Activations Density 0.174%

terms or phrases indicating classifications or categories

No Comments

No Known Activations