INDEX

Explanations

sentences addressing or referencing the reader or listener

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

Following "you" or "You"

np_acts-logits-general · gemini-2.0-flash

what you do or say

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 للمعارف

-0.72

makeConstraints

-0.70

حياته

-0.68

IntoConstraints

-0.61

oredCriteria

-0.59

ceği

-0.59

tvguidetime

-0.58

ficult

-0.57

extAlignment

-0.57

SpringBootTest

-0.57

POSITIVE LOGITS

 seem

0.69

 sound

0.69

 misunderstand

0.66

 seemed

0.64

 mean

0.63

 said

0.63

 forgot

0.63

 mentioned

0.62

 better

0.62

 sure

0.61

Activations Density 0.124%

sentences addressing or referencing the reader or listener

Following "you" or "You"

what you do or say

No Comments

No Known Activations