INDEX

Explanations

attends to mathematical terms or symbols of significance from their representations in surrounding context

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.13

1:0.12

2:0.11

3:0.09

4:0.06

5:0.03

6:0.07

7:0.35

Negative Logits

Personensuche

-0.32

Enllaces

-0.28

menuStrip

-0.28

 فريبيس

-0.28

multer

-0.27

 HasFactory

-0.27

 createSlice

-0.27

pid

-0.26

picasso

-0.26

entance

-0.25

POSITIVE LOGITS

 RSSSF

0.28

subsection

0.26

 Jehová

0.26

Ответ

0.25

MethodManager

0.25

 Vedi

0.25

IPAC

0.25

Референце

0.24

 JAXBElement

0.24

satunya

0.24

Activations Density 0.506%

attends to mathematical terms or symbols of significance from their representations in surrounding context

No Comments

No Known Activations