INDEX

Explanations

attends to connections between "and" tokens from names or relationships represented by the following tokens

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

0:0.07

1:0.10

2:0.08

3:0.12

4:0.09

5:0.04

6:0.26

7:0.20

Negative Logits

UNICIP

-0.29

 виправивши

-0.28

 kasarigan

-0.28

 RSSSF

-0.28

 gobernador

-0.28

Koordinaten

-0.28

นี่

-0.27

ടെ

-0.27

loroethane

-0.27

ookie

-0.27

POSITIVE LOGITS

IndentedString

0.35

enderror

0.33

usercontent

0.32

elemField

0.30

MethodManager

0.29

fromCharCode

0.28

GOTREF

0.28

mergeFrom

0.28

 useAppContext

0.27

WriteBarrier

0.27

Activations Density 0.012%

attends to connections between "and" tokens from names or relationships represented by the following tokens

No Comments

No Known Activations