INDEX

Explanations

period

np_max-act · gemini-2.0-flash

The neuron chiefly responds to the small decimal tokens that mark document formatting or indentation settings (e.g. “0.255126953125”) rather than ordinary words.

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

adh

-0.06

ADM

-0.06

-kind

-0.06

्रव

-0.06

_WRAP

-0.06

.flowLayoutPanel

-0.06

parcel

-0.06

 Sinn

-0.06

ци

-0.06

 adultes

-0.06

POSITIVE LOGITS

 Straight

0.07

aving

0.06

ệnh

0.06

ịnh

0.06

Beth

0.06

 Beth

0.06

.HandleFunc

0.06

========

0.06

 battled

0.06

 Virgin

0.06

Activations Density 0.020%

period

The neuron chiefly responds to the small decimal tokens that mark document formatting or indentation settings (e.g. “0.255126953125”) rather than ordinary words.

No Comments

No Known Activations