INDEX

Explanations

attends to dosage tokens marked "mg" from other similar markers

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

0:0.03

1:0.73

2:0.05

3:0.02

4:0.04

5:0.04

6:0.02

7:0.03

Negative Logits

Efq

-0.64

 ―――――

-0.60

 ་་

-0.59

 myſelf

-0.59

NUMX

-0.56

 Monfieur

-0.54

 itſelf

-0.54

^(@)

-0.54

 photolibrary

-0.52

 iſt

-0.52

POSITIVE LOGITS

0.42

0.38

for

0.35

in

0.35

is

0.34

to

0.34

0.33

Activations Density 2.682%

attends to dosage tokens marked "mg" from other similar markers

No Comments

No Known Activations