INDEX

Explanations

elements related to contradiction or negation in conjunction with positive reinforcement or supportive statements

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

preceding modal verbs

np_acts-logits-general · gemini-2.0-flash

enables and allows outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 AssemblyProduct

-0.55

GEBURTSDATUM

-0.54

 Temporada

-0.53

TypedDataSet

-0.53

WithMany

-0.50

Autoritní

-0.50

 procé

-0.48

 Fußballspieler

-0.48

 volon

-0.47

 Diſ

-0.47

POSITIVE LOGITS

 enables

0.81

 allows

0.79

 gives

0.78

 helps

0.73

 enable

0.73

 lead

0.68

 raises

0.66

 memudahkan

0.66

 provides

0.64

 allow

0.64

Activations Density 0.414%

elements related to contradiction or negation in conjunction with positive reinforcement or supportive statements

preceding modal verbs

enables and allows outcomes

No Comments

No Known Activations