INDEX

Explanations

candidate

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Patton

-0.07

plu

-0.06

experiment

-0.06

 rotor

-0.06

ีพ

-0.06

walls

-0.06

 initializes

-0.06

Gin

-0.05

 lạ

-0.05

(express

-0.05

POSITIVE LOGITS

 Candidates

0.10

 candidate

0.09

Candidates

0.08

Candidate

0.08

 candidates

0.08

anie

0.08

 Луч

0.07

 Candidate

0.07

arend

0.07

nant

0.07

Activations Density 0.011%

candidate

No Comments

No Known Activations