INDEX

Explanations

consuming

np_max-act · gemini-2.0-flash

The neuron activates on imperative consumption verbs (especially “eat” and “drink”) in advice lists.

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

 insulated

-0.07

²

-0.07

 take

-0.07

 replacing

-0.06

.handleChange

-0.06

ep

-0.06

 Investigators

-0.06

[ch

-0.06

 alterations

-0.06

_pci

-0.06

POSITIVE LOGITS

�

0.06

؟

0.06

 enormous

0.06

”?

0.06

언

0.06

 ।↵↵

0.06

Mario

0.06

 существ

0.06

.addCell

0.06

_REAL

0.06

Activations Density 0.018%

consuming

The neuron activates on imperative consumption verbs (especially “eat” and “drink”) in advice lists.

No Comments

No Known Activations