INDEX

Explanations

important

np_max-act · gemini-2.0-flash

This neuron fires on words that express strong significance or importance (e.g. “essential,” “vital,” “important”).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

丝

-0.07

 feasibility

-0.06

userInfo

-0.06

 footing

-0.06

 kcal

-0.06

textContent

-0.06

getReference

-0.06

Nun

-0.06

 Icons

-0.06

 Baldwin

-0.06

POSITIVE LOGITS

lendi

0.07

 setPage

0.07

이야

0.07

ographically

0.07

ecessary

0.07

ریه

0.06

DRV

0.06

ظمة

0.06

 مقدم

0.06

不断

0.06

Activations Density 0.043%

important

This neuron fires on words that express strong significance or importance (e.g. “essential,” “vital,” “important”).

No Comments

No Known Activations