INDEX

Explanations

Guid

np_max-act · gemini-2.0-flash

The neuron is strongly activated by GUID strings (hex‐digit sequences in curly braces) in MSBuild XML project files.

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

hydr

-0.06

(pipe

-0.06

 ขาย

-0.06

OUNTRY

-0.06

 fácil

-0.06

_street

-0.06

Empresa

-0.06

/chat

-0.06

_PLAYER

-0.06

ären

-0.06

POSITIVE LOGITS

��제

0.07

 impairment

0.07

]↵↵

0.07

Anc

0.07

 luận

0.06

 disqualified

0.06

实现

0.06

ượng

0.06

 viewHolder

0.06

meer

0.06

Activations Density 0.001%

Guid

The neuron is strongly activated by GUID strings (hex‐digit sequences in curly braces) in MSBuild XML project files.

No Comments

No Known Activations