INDEX

Explanations

key terms and identifiers related to options and parameters in a programming or technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

csol

-0.24

 fogo

-0.24

 financieros

-0.24

шенко

-0.23

 regionales

-0.23

ش

-0.23

 role

-0.23

dimento

-0.22

 prácti

-0.21

racuse

-0.21

POSITIVE LOGITS

RegressionTest

0.82

 Photocase

0.79

bootstrapcdn

0.70

Viited

0.69

astéroïdes

0.69

 pinulongan

0.68

TestingModule

0.68

 ویکی‌پدی

0.66



0.65

Kjelder

0.65

Activations Density 0.044%

key terms and identifiers related to options and parameters in a programming or technical context

No Comments

No Known Activations