INDEX

Explanations

logical arguments

np_max-act · gemini-2.0-flash

The neuron detects tokens that mark formal logical premises (e.g. “Premise,” numbered labels and their colons) in structured arguments.

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

.json

-0.08

-0.07

ized

-0.07

�

-0.06

 colorWith

-0.06

 Capture

-0.06

 frames

-0.06

 Behavior

-0.06

periences

-0.06

Step

-0.06

POSITIVE LOGITS

 thú

0.08

 termed

0.08

0.06

<!--↵

0.06

<TSource

0.06

 tiểu

0.06

 pode

0.06

 tük

0.06

ực

0.06

_Impl

0.06

Activations Density 0.023%

logical arguments

The neuron detects tokens that mark formal logical premises (e.g. “Premise,” numbered labels and their colons) in structured arguments.

No Comments

No Known Activations