INDEX

Explanations

mathematical or programming constructs related to coordinates and graphical representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

NameInMap

-0.63

 nonatomic

-0.58

]**

-0.57

ticoli

-0.54

 isInitialized

-0.54

tamanya

-0.53

########.

-0.53

 Réponses

-0.52

 BoxFit

-0.52

uced

-0.52

POSITIVE LOGITS

1.41

1.30

 plus

1.14

+(

1.13

$+

1.02

$+$

0.98

plus

0.98

+=

0.97

+\

0.96

＋

0.94

Activations Density 1.331%

mathematical or programming constructs related to coordinates and graphical representation

No Comments

No Known Activations