INDEX

Explanations

Probability letter replacement problems

np_max-act · gemini-2.0-flash

This neuron is sensitive to two‐letter sequences (bigrams) in the problem statements.

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

Modify

-0.06

 joking

-0.06

gray

-0.06

 ΑΠ

-0.06

ені

-0.06

�

-0.06

Ping

-0.06

aption

-0.06

 Fant

-0.06

.legend

-0.06

POSITIVE LOGITS

 demokrat

0.07

 delighted

0.07

0.06

	Returns

0.06

ुई

0.06

 genius

0.06

 agenda

0.06

 james

0.06

 میدان

0.06

Activations Density 0.004%

Probability letter replacement problems

This neuron is sensitive to two‐letter sequences (bigrams) in the problem statements.

No Comments

No Known Activations