INDEX

Explanations

attends to various types of tokens marked with "type" from class-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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.01

1:0.02

2:0.09

3:0.16

4:0.10

5:0.10

6:0.37

7:0.11

Negative Logits

 Monfieur

-1.02

 itſelf

-1.00

 Theſe

-0.95

 myſelf

-0.92

Efq

-0.91

 Majefty

-0.88

 ſeveral

-0.86

 ſche

-0.85

 ―――――

-0.85

 leaſt

-0.84

POSITIVE LOGITS

0.47

↵

0.44

0.43

<eos>

0.43

0.42

</strong>

0.40

↵↵↵

0.39

0.38

«

0.36

Activations Density 1.352%

attends to various types of tokens marked with "type" from class-related tokens

No Comments

No Known Activations