INDEX

Explanations

very or

np_max-act · gemini-2.0-flash

The neuron activates on degree-of-intensity words and qualifiers (intensifiers and hedges) that appear in quoted opinions or survey‐style responses.

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

 gaps

-0.07

 analyze

-0.07

Defense

-0.07

adı

-0.07

 totalCount

-0.07

 handling

-0.06

expand

-0.06

 сы

-0.06

 princess

-0.06

 asked

-0.06

POSITIVE LOGITS

�

0.07

.TypeString

0.07

'n

0.06

*pow

0.06

 wannonce

0.06

atos

0.06

.Prot

0.06

 freaking

0.06

(prom

0.06

Activations Density 0.009%

very or

The neuron activates on degree-of-intensity words and qualifiers (intensifiers and hedges) that appear in quoted opinions or survey‐style responses.

No Comments

No Known Activations