INDEX

Explanations

him

np_max-act · gemini-2.0-flash

This neuron flags occurrences of placeholder tokens used for character names (e.g. “NAME_1,” “NAME_2,” etc.).

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

त

-0.07

 Sharing

-0.07

 qualité

-0.07

 discarded

-0.07

WD

-0.07

umped

-0.06

 messages

-0.06

ayd

-0.06

ymi

-0.06

line

-0.06

POSITIVE LOGITS

 обеспеч

0.07

.minecraft

0.06

cx

0.06

HomeAsUpEnabled

0.06

 вида

0.06

_optimizer

0.06

 retr

0.06

	break

0.06

_filled

0.06

.optim

0.06

Activations Density 0.015%

him

This neuron flags occurrences of placeholder tokens used for character names (e.g. “NAME_1,” “NAME_2,” etc.).

No Comments

No Known Activations