INDEX

Explanations

phrases expressing correctness or validation of arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 particulière

-0.27

Portály

-0.25

 oral

-0.24

 evenings

-0.23

 Voll

-0.23

 Memorial

-0.23

 mito

-0.23

 routinely

-0.23

 frit

-0.22

lab

-0.22

POSITIVE LOGITS

Personendaten

0.77

Билгалдахарш

0.73

 rightfully

0.73

UnusedPrivate

0.71

 EconPapers

0.71

SequentialGroup

0.69

richTextPanel

0.66

PyExc

0.65

 rightly

0.65

fjspx

0.65

Activations Density 0.826%

phrases expressing correctness or validation of arguments

No Comments

No Known Activations

phrases expressing correctness or validation of arguments

No Comments

No Known Activations