INDEX

Explanations

code

np_max-act · gemini-2.0-flash

elements related to assembly metadata in code files.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

The neuron is specialized to spot GUID‐style hexadecimal identifier tokens (the hex digit groups and separators in those GUID strings).

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

с

-0.07

(KP

-0.07

्यकत

-0.07

(adj

-0.07

 domain

-0.07

 intents

-0.06

(keyword

-0.06

↵

-0.06

 detectors

-0.06

-cr

-0.06

POSITIVE LOGITS

[NUM

0.07

_raise

0.07

 Grü

0.06

-task

0.06

 troublesome

0.06

%%

0.06

	Logger

0.06

 hmot

0.06

_BATCH

0.06

meyen

0.06

Activations Density 0.004%

code

elements related to assembly metadata in code files.

The neuron is specialized to spot GUID‐style hexadecimal identifier tokens (the hex digit groups and separators in those GUID strings).

No Comments

No Known Activations