INDEX

Explanations

terms related to sensitivity or sensitivity measurement

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 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

 ويكيميديا

-0.99

BibitemShut

-0.99

principalTable

-0.92

 doubtnut

-0.89

CppMethod

-0.89

 ―――――

-0.88

extAlignment

-0.86

wiſe

-0.85

 otomatig

-0.83

 itſelf

-0.83

POSITIVE LOGITS

 span

0.78

 delicate

0.70

span

0.60

 winning

0.60

0.56

 sensitive

0.56

0.55

 Bond

0.53

<strong>

0.49

 spans

0.49

Activations Density 0.110%

terms related to sensitivity or sensitivity measurement

No Comments

No Known Activations