INDEX

Explanations

Informal conversations

np_max-act · gemini-2.0-flash

This neuron activates on structural special‐marker tokens (e.g. conversation header/footer tags) rather than on ordinary words.

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

 stimulus

-0.06

 speed

-0.06

 پیک

-0.06

 이해

-0.05

 ниже

-0.05

ラー

-0.05

_performance

-0.05

 medi

-0.05

 بزر

-0.05

�

-0.05

POSITIVE LOGITS

 Uber

0.08

 recib

0.07

 фер

0.07

 whilst

0.07

 гру

0.07

NRL

0.06

Gtk

0.06

 antlr

0.06

Navigator

0.06

Grupo

0.06

Activations Density 0.007%

Informal conversations

This neuron activates on structural special‐marker tokens (e.g. conversation header/footer tags) rather than on ordinary words.

No Comments

No Known Activations

Informal conversations

This neuron activates on structural special‐marker tokens (e.g. conversation header/footer tags) rather than on ordinary words.

No Comments

No Known Activations