INDEX

Explanations

d

np_max-act · gemini-2.0-flash

The neuron detects parts of legal case citations, in particular the “d” in reporter abbreviations (e.g. “F.3d,” “F.2d”) used in court opinions.

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

 مست

-0.07

_banner

-0.06

 Yani

-0.06

тра

-0.06

 frees

-0.06

_WRAPPER

-0.06

�

-0.06

�

-0.06

üçük

-0.06

ствен

-0.06

POSITIVE LOGITS

NSURLSession

0.07

_one

0.07

 cudd

0.06

dad

0.06

POP

0.06

-id

0.06

.ViewHolder

0.06

čky

0.06

}}"

0.06

-exclusive

0.06

Activations Density 0.001%

d

The neuron detects parts of legal case citations, in particular the “d” in reporter abbreviations (e.g. “F.3d,” “F.2d”) used in court opinions.

No Comments

No Known Activations