INDEX

Explanations

definitions and explanations from dictionaries

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

BASELINE

-0.50

vestream

-0.47

DoubleQuotes

-0.46

WithFormat

-0.46

ClientSize

-0.44

ParallelGroup

-0.41

ядок

-0.40

Testimonials

-0.39

 gepubliceerd

-0.39

ineo

-0.39

POSITIVE LOGITS

 dictionary

1.41

 definitions

1.39

 definition

1.33

 Dictionary

1.33

 Definitions

1.23

 Definition

1.22

 Merriam

1.22

 dictionaries

1.21

definition

1.15

dictionary

1.15

Activations Density 0.162%

definitions and explanations from dictionaries

No Comments

No Known Activations