INDEX

Explanations

references to subcategories and nested structures in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_14/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 ligiloj

-0.76

 CreateTagHelper

-0.59

cinogenicity

-0.59

łoży

-0.54

orianCalendar

-0.54

Odour

-0.53

 laço

-0.53

 ThemeData

-0.52

LEGGI

-0.52

actylus

-0.50

POSITIVE LOGITS

Sub

1.41

Sub

1.31

sub

1.29

SUB

1.27

sub

1.22

SUB

1.16

getSub

1.08

 getSub

1.08

 subs

0.91

サブ

0.86

Activations Density 0.122%

references to subcategories and nested structures in text

No Comments

No Known Activations