INDEX

Explanations

attends to printing-related tokens from other printing-related tokens

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

0:0.01

1:0.05

2:0.43

3:0.10

4:0.07

5:0.09

6:0.08

7:0.13

Negative Logits

Efq

-0.68

 itſelf

-0.64

 myſelf

-0.60

 Theſe

-0.57

 Jefus

-0.57

 Monfieur

-0.57

 becauſe

-0.55

 whoſe

-0.54

 Anſ

-0.54

^(@)

-0.54

POSITIVE LOGITS

<eos>

0.40

0.34

0.33

0.29

0.28

0.27

↵↵

0.27

Activations Density 0.383%

attends to printing-related tokens from other printing-related tokens

No Comments

No Known Activations