INDEX

Explanations

persists

np_max-act · gemini-2.0-flash

The neuron activates on words referring to problems or issues (e.g. “issue,” “problem,” “errors”).

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

-drop

-0.08

ιστη

-0.07

maps

-0.07

_TP

-0.07

 seni

-0.06

corn

-0.06

_SCENE

-0.06

 Coffee

-0.06

 Imagine

-0.06

 Drop

-0.06

POSITIVE LOGITS

 simplified

0.06

(robot

0.06

ignite

0.06

Beh

0.06

 vlastně

0.06

 červ

0.06

 verbally

0.06

 pilgr

0.06

 justifyContent

0.06

_atoms

0.06

Activations Density 0.008%

persists

The neuron activates on words referring to problems or issues (e.g. “issue,” “problem,” “errors”).

No Comments

No Known Activations

persists

The neuron activates on words referring to problems or issues (e.g. “issue,” “problem,” “errors”).

No Comments

No Known Activations