INDEX

Explanations

phrases and references related to commendations or endorsements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

帖最后由

-0.64

 springfox

-0.63

μβρίου

-0.62

RTLR

-0.62

EndGlobalSection

-0.61

hoeddwyd

-0.60

 архивлан

-0.59

曖昧さ回避

-0.58

Referanser

-0.58

afficheront

-0.57

POSITIVE LOGITS

]")]

0.63

 tempting

0.55

iteter

0.49

:]:

0.48

 appeals

0.48

 tempt

0.46

 appealing

0.45

うら

0.45

イ

0.45

 članak

0.44

Activations Density 0.376%

phrases and references related to commendations or endorsements

No Comments

No Known Activations