INDEX

Explanations

phrases and connections related to logic or logical connectors in textual arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

verwijspagina

-0.49

 instead

-0.47

Spoljašnje

-0.43

oa̍t

-0.41

Instead

-0.40

 betweenstory

-0.40

MockBean

-0.38

 resourceCulture

-0.37

expandindo

-0.37

 Instead

-0.36

POSITIVE LOGITS

 nakalista

0.68

complexContent

0.50

 Normdatei

0.46

orosi

0.45

 BoxFit

0.44

 ivelany

0.43

énario

0.42

 zelfs

0.42

 też

0.41

ValueGenerated

0.41

Activations Density 0.432%

phrases and connections related to logic or logical connectors in textual arguments

No Comments

No Known Activations