INDEX

Explanations

references to food accessibility and cultural highlights in the context of restaurants and dining experiences

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

Follows the word "find."

np_acts-logits-general · gemini-2.0-flash

will find or can be found

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

WriteTagHelper

-0.49

mergeFrom

-0.46

leur

-0.45

 Sache

-0.43

litt

-0.43

мся

-0.41

íticos

-0.40

Населення

-0.40

 opérateur

-0.40

itron

-0.39

POSITIVE LOGITS

 find

1.27

find

1.19

 finds

1.18

Find

1.14

 found

1.13

 Find

1.10

 Found

1.09

finds

1.08

 FIND

1.07

 finding

1.06

Activations Density 0.216%

references to food accessibility and cultural highlights in the context of restaurants and dining experiences

Follows the word "find."

will find or can be found

No Comments

No Known Activations