INDEX

Explanations

This neuron seems to be matching on a variety of different phrases, some related to parameters, some to threads and types, some to changes in state, and some XML tags consisting of bracketed characters, so it does not seem to have any particular purpose.

oai_token-act-pair · gemini-2.0-flash

code or markup

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 мәкал

-0.54

memoized

-0.51

 richtig

-0.49

resave

-0.48

 sbag

-0.47

 ComVisible

-0.46

::~

-0.46

Cod

-0.43

Ech

-0.43

lords

-0.43

POSITIVE LOGITS

 للمعارف

0.56

Географи

0.54

 lenker

0.53

httphttps

0.52

__":

0.52

 currency

0.52

 الحره

0.52

 במח

0.52

 Stapleton

0.50

AxisAlignment

0.49

Activations Density 0.019%

This neuron seems to be matching on a variety of different phrases, some related to parameters, some to threads and types, some to changes in state, and some XML tags consisting of bracketed characters, so it does not seem to have any particular purpose.

code or markup

No Comments

No Known Activations

This neuron seems to be matching on a variety of different phrases, some related to parameters, some to threads and types, some to changes in state, and some XML tags consisting of bracketed characters, so it does not seem to have any particular purpose.

code or markup

No Comments

No Known Activations