INDEX

Explanations

Code and programming

np_max-act · gemini-2.0-flash

programming-related configurations and metadata.

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

This neuron activates on identifier tokens that are names of configuration or template parameters (e.g. the dotted or underscored keys inside `${…}` placeholders).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

_physical

-0.06

ISI

-0.06

 kısa

-0.06

(LL

-0.06

อด

-0.06

ł

-0.06

 Valor

-0.06

 blockade

-0.06

_math

-0.06

 Murdoch

-0.06

POSITIVE LOGITS

 Responsible

0.07

 Assuming

0.07

 Παρ

0.07

SEX

0.06

.Time

0.06

	system

0.06

 baking

0.06

 Allied

0.06

_MAP

0.06

ايت

0.06

Activations Density 0.050%

Code and programming

programming-related configurations and metadata.

This neuron activates on identifier tokens that are names of configuration or template parameters (e.g. the dotted or underscored keys inside `${…}` placeholders).

No Comments

No Known Activations