INDEX

Explanations

Code

np_max-act · gemini-2.0-flash

This neuron detects member‐accesses on underlying/delegate fields in wrapper classes (e.g. “wrapped.”, “delegate.”, “innerTracer.”, “_operation.”).

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

IAM

-0.07

 인간

-0.06

 Lunar

-0.06

้ำหน

-0.06

*z

-0.06

сен

-0.06

松

-0.06

orrar

-0.06

esar

-0.06

냐

-0.06

POSITIVE LOGITS

 Something

0.07

 زبان

0.07

bob

0.06

(Py

0.06

IFICATE

0.06

UITableView

0.06

Nie

0.06

vous

0.06

	super

0.06

 tert

0.06

Activations Density 0.014%

Code

This neuron detects member‐accesses on underlying/delegate fields in wrapper classes (e.g. “wrapped.”, “delegate.”, “innerTracer.”, “_operation.”).

No Comments

No Known Activations