INDEX

Explanations

questions and references about understanding and discussing various topics

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

Questions using "what," "is," or "are."

np_acts-logits-general · gemini-2.0-flash

what it's all about

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

 Jefus

-0.65

 pleaſure

-0.65

ViewFeatures

-0.64

 myſelf

-0.61

 reaſon

-0.61

 Monfieur

-0.60

 ſame

-0.60

 betweenstory

-0.60

xase

-0.60

 Reſ

-0.59

POSITIVE LOGITS

 CreateTagHelper

0.61

AnimationsModule

0.56

 based

0.54

MetaObject

0.53

 entails

0.53

 entail

0.53

 consist

0.52

InstrumentedTest

0.52

 entailed

0.51

 exactly

0.50

Activations Density 0.138%

questions and references about understanding and discussing various topics

Questions using "what," "is," or "are."

what it's all about

No Comments

No Known Activations