INDEX

Explanations

infections in immunocompromised

np_max-act · gemini-2.0-flash

The neuron activates on subword fragments of “immunocompromised” or “immunocompetent,” i.e. mentions of a patient’s immune‐compromised status.

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

.byte

-0.07

pair

-0.06

دة

-0.06

ante

-0.06

�

-0.06

 پنج

-0.06

ili

-0.06

yyy

-0.06

دث

-0.06

ys

-0.06

POSITIVE LOGITS

ocom

0.08

(Un

0.07

 footing

0.07

 stocked

0.07

.Conn

0.07

 boosts

0.07

 Disqus

0.06

官方

0.06

 Unicode

0.06

งเศส

0.06

Activations Density 0.001%

infections in immunocompromised

The neuron activates on subword fragments of “immunocompromised” or “immunocompetent,” i.e. mentions of a patient’s immune‐compromised status.

No Comments

No Known Activations