INDEX

Explanations

specific identifiers or keywords related to programming, processes, or searching functionalities

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

endphp

-0.42

ddots

-0.42

corrhi

-0.40

styleUrls

-0.36

もちゃ

-0.36

iterranée

-0.35

 useContext

-0.34

しゃれ

-0.34

odyne

-0.34

olato

-0.33

POSITIVE LOGITS

LookAnd

0.57

Cyfeiriadau

0.55

 favourite

0.51

 favorite

0.50

 OFDb

0.49

favorite

0.47

favourite

0.45

UpInside

0.45

 favorites

0.44

AsUp

0.43

Activations Density 0.119%

specific identifiers or keywords related to programming, processes, or searching functionalities

No Comments

No Known Activations