INDEX

Explanations

terms related to various states and physical objects, particularly in contexts involving positions or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/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.15.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

awtextra

-0.81

 EconPapers

-0.72

webElement

-0.71

 snippetHide

-0.71

calciatore

-0.67

хьтан

-0.64

ORAGE

-0.63

kháu

-0.62

 surla

-0.62

affaire

-0.62

POSITIVE LOGITS

 asti

0.59

までの

0.59

via

0.56

まで

0.51

へ

0.50

 straight

0.47

Via

0.47

까지

0.46

SizeF

0.44

straight

0.44

Activations Density 0.387%

terms related to various states and physical objects, particularly in contexts involving positions or locations

No Comments

No Known Activations