INDEX

Explanations

instances of the ampersand symbol (&) in various formats and contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

numbers and code symbols

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+#

-0.81

OGND

-0.73

SequentialGroup

-0.66

ChildScrollView

-0.66

expandindo

-0.61

 henvisninger

-0.59

fjspx

-0.59

addPreferredGap

-0.59

もしれません

-0.58

にほんブログ村

-0.58

POSITIVE LOGITS

0.43

<td>

0.43

</td>

0.42

<th>

0.37

int

0.35

&=&

0.35

<h1>

0.35

//

0.35

main

0.34

Activations Density 0.018%

instances of the ampersand symbol (&) in various formats and contexts

numbers and code symbols

No Comments

No Known Activations