INDEX

Explanations

le

np_max-act · gemini-2.0-flash

The neuron detects the “pre-” prefix at the start of words (e.g. predisposing terms like prediabetes, pre‐phase, preleukemia, prefrailty).

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

پر

-0.07

daemon

-0.06

lename

-0.06

_EXP

-0.06

Sector

-0.06

 apenas

-0.06

ाय

-0.06

CET

-0.06

每

-0.06

 fists

-0.06

POSITIVE LOGITS

_LAYOUT

0.08

キャ

0.07

 preprocess

0.07

OT

0.07

브

0.07

 precursor

0.07

/************************************************

0.07

사람

0.07

 datab

0.06

κι

0.06

Activations Density 0.028%

le

The neuron detects the “pre-” prefix at the start of words (e.g. predisposing terms like prediabetes, pre‐phase, preleukemia, prefrailty).

No Comments

No Known Activations