INDEX

Explanations

to

np_max-act · gemini-2.0-flash

The neuron is peaking on temporal or scheduling expressions—tokens that specify when or how often (e.g. “Twice monthly,” “every day,” “set aside time”) in instructional contexts.

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

보증금

-0.07

ago

-0.06

 Wednesday

-0.06

_poly

-0.06

 clar

-0.06

iddet

-0.06

 cooking

-0.06

 initial

-0.06

 Friday

-0.06

 Sunday

-0.06

POSITIVE LOGITS

 контр

0.07

 richt

0.06

hapus

0.06

 ayrıntı

0.06

�

0.06

ター

0.06

мах

0.06

$res

0.06

_supported

0.06

ινό

0.06

Activations Density 0.075%

to

The neuron is peaking on temporal or scheduling expressions—tokens that specify when or how often (e.g. “Twice monthly,” “every day,” “set aside time”) in instructional contexts.

No Comments

No Known Activations