INDEX

Explanations

structured numerical information, likely related to scientific or academic references

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

Followed by punctuation in parentheses

np_acts-logits-general · gemini-2.0-flash

10

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

XmlAccessType

-0.67

:✨

-0.65

ImageContext

-0.56

 morire

-0.55

 esquerda

-0.53

 internacionais

-0.52

 opportunità

-0.51

 trần

-0.50

indl

-0.50

 gomma

-0.50

POSITIVE LOGITS

 الرياضيه

0.74

 adjourned

0.57

BagConstraints

0.56

 actionPerformed

0.56

 szóci

0.55

findpost

0.54

 comp

0.54

Бахар

0.54

theless

0.53

InjectAttribute

0.53

Activations Density 0.207%

structured numerical information, likely related to scientific or academic references

Followed by punctuation in parentheses

10

No Comments

No Known Activations