INDEX

Explanations

references to academic publications and conferences

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

<?

-0.47

ウンテン

-0.46

SuspendLayout

-0.46

dem

-0.45

 dicendo

-0.43

loj

-0.42

前は

-0.42

 verbally

-0.42

++)

-0.41

bota

-0.41

POSITIVE LOGITS

 Proceedings

0.87

 proceedings

0.86

 EconPapers

0.86

Proceedings

0.84

 PROCEEDINGS

0.82

 lenker

0.81

 Meksiku

0.79

Datuak

0.76

 BoxDecoration

0.75

 Fests

0.74

Activations Density 0.226%

references to academic publications and conferences

No Comments

No Known Activations