INDEX

Explanations

Questions and answers

np_max-act · gemini-2.0-flash

the start of an assistant’s reply in chat-style message formatting (assistant role/boundary markers).

oai_token-act-pair · gpt-5 Triggered by @vetterc0

New Auto-Interp

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Org

-0.07

ends

-0.07

Steve

-0.07

texture

-0.07

すぐに

-0.07

mt

-0.06

 Arctic

-0.06

这意味着

-0.06

 BaseService

-0.06

sk

-0.06

POSITIVE LOGITS

 mism

0.07

изации

0.07

قوم

0.07

畛

0.07

acional

0.07

emploi

0.07

_FILL

0.07

обор

0.06

Votes

0.06

_BIN

0.06

Activations Density 0.053%

Questions and answers

the start of an assistant’s reply in chat-style message formatting (assistant role/boundary markers).

No Comments

No Known Activations