INDEX

Explanations

numerical references and quantifications within a text

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

MethodManager

-0.50

出版年

-0.48

GIVEREF

-0.48

writeFieldEnd

-0.44

BagConstraints

-0.41

 springfox

-0.41

 Fundamental

-0.40

 linkovi

-0.40

 inca

-0.39

AddTagHelper

-0.39

POSITIVE LOGITS

 historien

0.52

 historical

0.48

 historiador

0.45

 history

0.45

 sejarah

0.44

 colorés

0.44

 asiática

0.44

 historians

0.43

histoire

0.43

TagMode

0.42

Activations Density 0.089%

numerical references and quantifications within a text

No Comments

No Known Activations