INDEX

Explanations

phrases that indicate a process or procedure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 kasarigan

-0.72

tvguidetime

-0.72

afficheront

-0.67

 ComVisible

-0.61

 cherchés

-0.58

xase

-0.56

 propOrder

-0.56

 okuyayım

-0.55

GOTREF

-0.54

lectricité

-0.53

POSITIVE LOGITS

 ContentValues

0.64

SpringRunner

0.58

 Altern

0.58

 Note

0.57

Altern

0.57

altern

0.53

Note

0.51

repe

0.50

 apologise

0.49

FieldNumber

0.49

Activations Density 0.453%

phrases that indicate a process or procedure

No Comments

No Known Activations