INDEX

Explanations

attends to numeric tokens surrounded by commas from other tokens in the sequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

0:0.49

1:0.13

2:0.07

3:0.05

4:0.06

5:0.07

6:0.04

7:0.05

Negative Logits

 themſelves

-0.52

 myſelf

-0.50

 himſelf

-0.48

 ſche

-0.47

expandindo

-0.46

 houſe

-0.45

 ſtate

-0.45

 فريبيس

-0.45

 ſta

-0.45

 समीक्षाओं

-0.45

POSITIVE LOGITS

ftrag

0.21

 Espinosa

0.21

gross

0.20

 Author

0.20

olulu

0.20

0.19

<unused61>

0.19

<unused62>

0.19

 služ

0.19

pun

0.19

Activations Density 1.055%

attends to numeric tokens surrounded by commas from other tokens in the sequence

No Comments

No Known Activations