INDEX

Explanations

mathematical concepts and terminology related to convergence rates in function approximation

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

BeginContext

-0.56

WebServlet

-0.51

ihnachts

-0.48

biet

-0.47

 despe

-0.46

Grat

-0.45



-0.45

msgTypes

-0.45

MEMORANDUM

-0.45

NegativeButton

-0.44

POSITIVE LOGITS

 norm

0.70

 censiti

0.69

 norms

0.64

twimg

0.60

Norm

0.58

 NORM

0.58

 Relaxed

0.58

norm

0.58

infty

0.56

 norme

0.55

Activations Density 1.036%

mathematical concepts and terminology related to convergence rates in function approximation

No Comments

No Known Activations