INDEX

Explanations

attends to the token "doi" from citation-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

0:0.02

1:0.03

2:0.03

3:0.68

4:0.04

5:0.03

6:0.08

7:0.05

Negative Logits

 myſelf

-0.49

 itſelf

-0.40

ſelves

-0.40

 Shakspeare

-0.39

 Monfieur

-0.38

 reaſon

-0.38

 purpoſe

-0.37

 Cæsar

-0.37

 becauſe

-0.37

 raiſ

-0.37

POSITIVE LOGITS

De

0.37

Tag

0.35

Har

0.34

Mag

0.34

Az

0.33

Cas

0.33

Am

0.32

 gran

0.32

Gar

0.31

Sin

0.31

Activations Density 0.741%

attends to the token "doi" from citation-related tokens

No Comments

No Known Activations