INDEX

Explanations

evidence of decision-making or evaluation in 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 @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 мәкал

-0.57

NameInMap

-0.54

 defaultstate

-0.53

@"/

-0.43

ActionEvent

-0.42

</h6>

-0.42

NamedQueries

-0.41

ItemBackground

-0.40

 <>",

-0.40

 Signalez

-0.40

POSITIVE LOGITS

tagHelperRunner

0.68

UnusedPrivate

0.63

parsedMessage

0.57

enterOuterAlt

0.54

 HasFactory

0.54

 &___

0.53

 iconLine

0.52

requireNonNull

0.46

 ProtoMessage

0.45

 dentaire

0.45

Activations Density 0.739%

evidence of decision-making or evaluation in a text

No Comments

No Known Activations