INDEX

Explanations

attends to various noun-related tokens from general descriptive terms later in the sequence

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

1:0.18

2:0.09

3:0.07

4:0.16

5:0.03

6:0.09

7:0.20

Negative Logits

-0.30

-0.26

 режиме

-0.26

-0.25

-0.24

näm

-0.23

 来自

-0.23

vata

-0.22

POSITIVE LOGITS

LookAnd

0.51

 виправивши

0.49

 استنادى

0.46

ſelves

0.43

 tdci

0.43

ništvo

0.43

Personensuche

0.43

rungsseite

0.42

 vectorstock

0.42

 NSCoder

0.41

Activations Density 0.471%

attends to various noun-related tokens from general descriptive terms later in the sequence

No Comments

No Known Activations