INDEX

Explanations

phrases that indicate expert opinions and widespread observations on various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

 '\\;'

-0.52

AutoScale

-0.47

SPATH

-0.46

verifyException

-0.45

 queſta

-0.44

(;;)

-0.43

SequentialGroup

-0.43

ᅠ

-0.42

 EconPapers

-0.42

 intptr

-0.42

POSITIVE LOGITS

 debated

0.45

 רבים

0.43

 говорят

0.43

 theorists

0.43

 touted

0.42

 critics

0.42

 Anerkennung

0.42

 lluvia

0.41

 widely

0.40

 Experten

0.40

Activations Density 0.660%

phrases that indicate expert opinions and widespread observations on various topics

No Comments

No Known Activations