INDEX

Explanations

references to academic conferences or publications related to artificial intelligence and machine learning

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

:✨

-0.58

endpush

-0.56

℗

-0.53

AddTagHelper

-0.52

twimg

-0.52

usura

-0.52

 élevées

-0.52

criture

-0.52

fjspx

-0.50

 élevés

-0.50

POSITIVE LOGITS

 Computing

0.68

Computing

0.63

 оригинала

0.55

computing

0.53

stuhl

0.52

 deleteUser

0.51

Jude

0.50

 Computers

0.49

otea

0.49

 Technolog

0.49

Activations Density 0.459%

references to academic conferences or publications related to artificial intelligence and machine learning

No Comments

No Known Activations