INDEX

Explanations

terms and concepts related to software licensing and warranties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 aici

-0.50

 habet

-0.49

 efficacité

-0.48

 παιδιά

-0.48

 Italij

-0.48

 vorbei

-0.47

 viata

-0.47

 ervaren

-0.47

 kvinder

-0.47

 quit

-0.47

POSITIVE LOGITS

 виправивши

1.16

دانشنامهٔ

0.92

 дописавши

0.90

 ویکی‌پدیای

0.89

0.86

 Roskov

0.85

 typelib

0.82

SequentialGroup

0.80

sidemargin

0.80

InputTagHelper

0.79

Activations Density 0.036%

terms and concepts related to software licensing and warranties

No Comments

No Known Activations