INDEX

Explanations

instances of the word "and" indicating connections or accumulation in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 réalis

-0.56

PerformLayout

-0.56

 CreateTagHelper

-0.56

Tại

-0.54

 fulfill

-0.53

 defy

-0.53

 steal

-0.52

 renovate

-0.52

最快更新

-0.51

itemBuilder

-0.50

POSITIVE LOGITS

gorithm

0.67

werd

0.66

帖最后由

0.65

EDEFAULT

0.64

atever

0.63

nteral

0.62

PUBL

0.61

".$_

0.61

 zostanie

0.60

 boards

0.60

Activations Density 0.331%

instances of the word "and" indicating connections or accumulation in the text

No Comments

No Known Activations