INDEX

Explanations

attends to factors marked earlier from various categories of factors marked later

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

0:0.03

1:0.52

2:0.09

3:0.03

4:0.10

5:0.11

6:0.03

7:0.06

Negative Logits

-0.67

-0.65

-0.54

-0.48

↵

-0.47

</em>

-0.45

</i>

-0.44

is

-0.44

POSITIVE LOGITS

oredCriteria

0.96

 myſelf

0.92

 itſelf

0.91

Efq

0.87

Personendaten

0.85

IsMutable

0.84

 ―――――

0.84

 Мексичка

0.84

 للمعارف

0.84

 ExecuteAsync

0.83

Activations Density 0.054%

attends to factors marked earlier from various categories of factors marked later

No Comments

No Known Activations