INDEX

Explanations

prepositional phrases and their modifiers

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

from other languages and prepositional phrases

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

TagMode

-0.69

 незавершена

-0.63

 Cæsar

-0.63

elemField

-0.60

kháu

-0.60

 myſelf

-0.60

parsedMessage

-0.59

 iprot

-0.59

Бахар

-0.58

 queſta

-0.58

POSITIVE LOGITS

 reported

0.36

Reported

0.30

 Reported

0.30

 report

0.30

굽

0.29

制

0.29

ἑ

0.29

Auf

0.29

Though

0.29

sser

0.29

Activations Density 0.020%

prepositional phrases and their modifiers

from other languages and prepositional phrases

No Comments

No Known Activations