INDEX

Explanations

programming-related terms and coding syntax

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

 boca

-0.50

 الحره

-0.50

치

-0.47

potamus

-0.44

-0.43

 Direct

-0.43

 Ross

-0.42

 réfrigérateur

-0.42

 frigor

-0.41

sedia

-0.41

POSITIVE LOGITS

 cherchés

0.82

 Roskov

0.81

 فريبيس

0.80

 يتيمه

0.80

 Monfieur

0.78

WEBPACK

0.77

 webpack

0.77

 contextLoads

0.76

 loaders

0.70

DockStyle

0.70

Activations Density 0.094%

programming-related terms and coding syntax

No Comments

No Known Activations