INDEX

Explanations

Code snippets

np_max-act · gemini-2.0-flash

This neuron detects the special “[� � �CLASSIC]” bracketed tag sequence used to mark the standard-response classifier token.

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

 alertDialog

-0.06

 immature

-0.06

 parch

-0.06

 initialization

-0.06

 concat

-0.06

 Mods

-0.06

weapon

-0.06

 missionary

-0.06

 inconsist

-0.06

 Wars

-0.06

POSITIVE LOGITS

 Localization

0.07

(view

0.07

首页

0.07

 licz

0.07

.Once

0.07

enef

0.06

Tile

0.06

coeff

0.06

.ForEach

0.06

 routes

0.06

Activations Density 0.002%

Code snippets

This neuron detects the special “[� � �CLASSIC]” bracketed tag sequence used to mark the standard-response classifier token.

No Comments

No Known Activations