INDEX

Explanations

:

np_max-act · gemini-2.0-flash

Tokens that mark the assistant/response header and related system metadata (i.e., assistant role and start/end header tokens).

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

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-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

 layout

-0.07

 разд

-0.07

counter

-0.06

opc

-0.06

.Al

-0.06

.NoSuch

-0.06

त

-0.06

Daw

-0.06

�

-0.06

 reporting

-0.06

POSITIVE LOGITS

 كبير

0.07

 biệt

0.07

icies

0.06

 gunshot

0.06

γων

0.06

 през

0.06

anut

0.06

 Made

0.06

ій

0.06

izon

0.06

Activations Density 0.093%

:

Tokens that mark the assistant/response header and related system metadata (i.e., assistant role and start/end header tokens).

No Comments

No Known Activations

:

Tokens that mark the assistant/response header and related system metadata (i.e., assistant role and start/end header tokens).

No Comments

No Known Activations