INDEX

Explanations

new to doing something

np_max-act · gemini-2.0-flash

The neuron is detecting first‐person personal narrative tokens—auxiliary verbs, participles, and adverbs (e.g. “been,” “using,” “learning,” “just,” “sorry”) that speakers use when describing their own experience.

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

 форма

-0.07

@NoArgsConstructor

-0.06

 secondo

-0.06

前の

-0.06

 обычно

-0.06

 هنوز

-0.06

 бути

-0.05

 residuals

-0.05

рем

-0.05

たら

-0.05

POSITIVE LOGITS

 พร

0.08

soo

0.08

Ў

0.07

VENTORY

0.07

typing

0.07

 Hawaiian

0.07

LLU

0.07

PRI

0.06

 Proceedings

0.06

 Πανεπ

0.06

Activations Density 0.065%

new to doing something

The neuron is detecting first‐person personal narrative tokens—auxiliary verbs, participles, and adverbs (e.g. “been,” “using,” “learning,” “just,” “sorry”) that speakers use when describing their own experience.

No Comments

No Known Activations

new to doing something

The neuron is detecting first‐person personal narrative tokens—auxiliary verbs, participles, and adverbs (e.g. “been,” “using,” “learning,” “just,” “sorry”) that speakers use when describing their own experience.

No Comments

No Known Activations