INDEX

Explanations

words associated with positive evaluations or quality

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

"Good" followed by a noun

np_acts-logits-general · gemini-2.0-flash

Good contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

tvguidetime

-0.68

DockStyle

-0.66

tlement

-0.62

DebuggerNonUser

-0.62

InjectAttribute

-0.60

tonsoft

-0.60

eradish

-0.57

XmlAccessorType

-0.56

 viewType

-0.56

querySelectorAll

-0.56

POSITIVE LOGITS

byes

0.91

 samar

0.88

 Samaritan

0.84

enough

0.81

 luck

0.81

 Samar

0.79

 fortune

0.78

 deeds

0.77

ol

0.76

night

0.74

Activations Density 0.102%

words associated with positive evaluations or quality

"Good" followed by a noun

Good contexts

No Comments

No Known Activations