INDEX

Explanations

questions or comments regarding users' actions and experiences

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

Follows special characters or abbreviations

np_acts-logits-general · gemini-2.0-flash

programming language constructs

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

 fhew

-0.77

 poffible

-0.71

Efq

-0.68

ſelf

-0.66

 myſelf

-0.66

mêmes

-0.66

 pleaſure

-0.65

 chofe

-0.65

 raiſ

-0.65

 auffi

-0.62

POSITIVE LOGITS

 propOrder

0.95

InjectAttribute

0.80

FunctionFlags

0.77

 ujednoznacz

0.76

 EconPapers

0.75

rungsseite

0.73

ISupport

0.69

SequentialGroup

0.69

UserScript

0.68

featureID

0.67

Activations Density 0.124%

questions or comments regarding users' actions and experiences

Follows special characters or abbreviations

programming language constructs

No Comments

No Known Activations