INDEX

Explanations

code specifying size limits

np_max-act · gemini-2.0-flash

The neuron responds to numeric literals (and their surrounding punctuation) in code—especially when they appear as parameter values like “maxResults: 10.”

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

BracketAccess

-0.07

(REG

-0.07

_COMBO

-0.07

 tuples

-0.07

 intf

-0.07

 byste

-0.07

駅

-0.06

 bychom

-0.06

 Roths

-0.06

	HRESULT

-0.06

POSITIVE LOGITS

 unpredict

0.07

 prox

0.06

 anch

0.06

 velocidad

0.06

ersist

0.06

/new

0.06

ляти

0.06

 omas

0.06

reamble

0.06

�

0.06

Activations Density 0.014%

code specifying size limits

The neuron responds to numeric literals (and their surrounding punctuation) in code—especially when they appear as parameter values like “maxResults: 10.”

No Comments

No Known Activations