INDEX

Explanations

references to words and language

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

"word" or similar variations

np_acts-logits-general · gemini-2.0-flash

word followed by other words

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

‍♂️

-0.61

 betrek

-0.55

]-->

-0.54

 يتيمه

-0.53

 Shreve

-0.53

πως

-0.52

FileChooser

-0.50

 Shun

-0.50

UnitTesting

-0.50

 Normdatei

-0.49

POSITIVE LOGITS

 Word

0.85

 WORD

0.83

 Words

0.82

 words

0.79

 WORDS

0.78

Word

0.72

 word

0.70

Words

0.67

 sumpay

0.67

 հղումներ

0.64

Activations Density 0.218%

references to words and language

"word" or similar variations

word followed by other words

No Comments

No Known Activations