INDEX

Explanations

words related to programming and structured data handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

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

Few

-0.44

 tahu

-0.42

getKeyCode

-0.41

 précie

-0.41

 milliards

-0.41

 millions

-0.40

 hombros

-0.40

اريخ

-0.38

 impegn

-0.37

millions

-0.37

POSITIVE LOGITS

 ***!

0.87

 MenuView

0.84

 CreateTagHelper

0.81

LookAnd

0.77

httphttps

0.76

GEBURTSDATUM

0.75

 ſever

0.72

UrlResolution

0.72

PerformLayout

0.71

 تانيه

0.70

Activations Density 0.013%

words related to programming and structured data handling

No Comments

No Known Activations