INDEX

Explanations

comparative statements and phrases indicating performance or condition in relation to others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

exitRule

-0.46

awtextra

-0.44

InstrumentedTest

-0.41

 fusca

-0.38

idase

-0.37

 endwhile

-0.36

 giang

-0.36

ENSUS

-0.36

getMenuInflater

-0.36

Separ

-0.35

POSITIVE LOGITS

 referrerpolicy

0.50

 EconPapers

0.46

HostException

0.44

 kaybet

0.43

]")]

0.41

hematics

0.41

帖最后由

0.40

󠁣

0.40

enderror

0.40

くら

0.40

Activations Density 0.096%

comparative statements and phrases indicating performance or condition in relation to others

No Comments

No Known Activations