INDEX

Explanations

attends from tokens that contain multiple non-space characters to subsequent tokens containing spaces

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.06

2:0.03

3:0.17

4:0.08

5:0.40

6:0.09

7:0.10

Negative Logits

-4.16

-1.95

-1.80

-1.70

-1.65

-1.59

and

-1.54

-1.53

-1.47

POSITIVE LOGITS

 myſelf

3.52

 itſelf

3.31

Efq

3.20

 Monfieur

3.09

 ―――――

3.09

 Theſe

3.03

 ་་

2.98

 Jefus

2.95

 Anſ

2.94

 Majefty

2.94

Activations Density 0.062%

attends from tokens that contain multiple non-space characters to subsequent tokens containing spaces

No Comments

No Known Activations