INDEX

Explanations

evaluative language related to quality or value, especially in terms of expectations and outcomes

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

Negation or uncertainty

np_acts-logits-general · gemini-2.0-flash

not perfectly clear

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

 onFailure

-0.56

 Paglinawan

-0.53

albeit

-0.52

்கள

-0.51

 safely

-0.50

 birlikte

-0.49

↘

-0.49

 Safely

-0.48

 responsibly

-0.47

bcryptjs

-0.47

POSITIVE LOGITS

 fancy

0.78

 huge

0.75

 perfectos

0.73

 jätte

0.70

 exact

0.69

 miracles

0.69

 massive

0.68

 flashy

0.67

 extensive

0.67

 massively

0.66

Activations Density 0.279%

evaluative language related to quality or value, especially in terms of expectations and outcomes

Negation or uncertainty

not perfectly clear

No Comments

No Known Activations