INDEX

Explanations

Answers

np_max-act · gemini-2.0-flash

This neuron detects empathetic opening phrases that begin with “It” (e.g., “It can be…,” “It sounds like…”) used to acknowledge and validate the user’s feelings.

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

й

-0.07

ůvodu

-0.07

�

-0.06

انگ

-0.06

亜

-0.06

 adultes

-0.06

asures

-0.06

ул

-0.06

 >>↵↵

-0.06

อร

-0.06

POSITIVE LOGITS

MPI

0.06

_sidebar

0.06

?",

0.06

 العمل

0.06

(dec

0.06

 základ

0.06

(props

0.06

:T

0.06

(FLAGS

0.06

ridged

0.06

Activations Density 0.100%

Answers

This neuron detects empathetic opening phrases that begin with “It” (e.g., “It can be…,” “It sounds like…”) used to acknowledge and validate the user’s feelings.

No Comments

No Known Activations

Answers

This neuron detects empathetic opening phrases that begin with “It” (e.g., “It can be…,” “It sounds like…”) used to acknowledge and validate the user’s feelings.

No Comments

No Known Activations