INDEX

Explanations

,

np_max-act · gemini-2.0-flash

This neuron detects hedging or subjective discourse markers (e.g. “but,” “I think,” “safe to say,” “well,” “something about”) that signal personal commentary or transition.

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

BG

-0.07

 mọi

-0.06

.wind

-0.06

ادی

-0.06

.EditValue

-0.06

 Explos

-0.06

ูก

-0.06

раниц

-0.06

 Před

-0.06

fan

-0.06

POSITIVE LOGITS

INCLUDED

0.06

SEQU

0.06

Й

0.06

fortunately

0.06

 TRANS

0.06

smtp

0.06

ForSegue

0.06

档

0.06

 adverts

0.06

ardu

0.06

Activations Density 0.059%

,

This neuron detects hedging or subjective discourse markers (e.g. “but,” “I think,” “safe to say,” “well,” “something about”) that signal personal commentary or transition.

No Comments

No Known Activations