INDEX

Explanations

code

np_max-act · gemini-2.0-flash

This neuron detects tokens that are part of function or method call names (API or library call identifiers) in code.

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

 exporter

-0.07

assel

-0.06

Candidates

-0.06

keletal

-0.06

IFF

-0.06

Kill

-0.06

*T

-0.06

.inverse

-0.06

apo

-0.06

 diplomat

-0.06

POSITIVE LOGITS

 خانه

0.07

 beer

0.06

 tavsiye

0.06

 piercing

0.06

 résult

0.06

 데이터

0.06

vault

0.06

uç

0.06

.entrySet

0.06

改

0.06

Activations Density 0.170%

code

This neuron detects tokens that are part of function or method call names (API or library call identifiers) in code.

No Comments

No Known Activations

code

This neuron detects tokens that are part of function or method call names (API or library call identifiers) in code.

No Comments

No Known Activations