INDEX

Explanations

questions directed towards the reader, often beginning with "do you" or similar phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 ""),

-0.71

TypedDataSet

-0.69

 fevere

-0.65

تقاوى

-0.63

 poffe

-0.63

 Portail

-0.63

 llev

-0.62

'><

-0.61

 []),

-0.61

 ")";

-0.61

POSITIVE LOGITS

 Duckworth

0.79

DockStyle

0.75

AnimationsModule

0.72

 Graff

0.70

toHaveBeenCalled

0.69

iserie

0.67

knex

0.66

 MainAxisSize

0.65

Mã

0.64

 Budget

0.64

Activations Density 0.096%

questions directed towards the reader, often beginning with "do you" or similar phrases

No Comments

No Known Activations