INDEX

Explanations

attends to tokens marked "for" from various related phrases or clauses preceding them

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

0:0.04

1:0.12

2:0.46

3:0.08

4:0.07

5:0.08

6:0.06

7:0.06

Negative Logits

 myſelf

-5.50

Efq

-5.44

 itſelf

-5.22

 Monfieur

-4.94

 Theſe

-4.94

^(@)

-4.69

 ―――――

-4.69

 Jefus

-4.63

 Majefty

-4.56

 themſelves

-4.53

POSITIVE LOGITS

2.88

2.59

2.58

2.50

↵↵

2.48

2.31

↵

2.27

<eos>

2.27

2.19

in

2.16

Activations Density 1.785%

attends to tokens marked "for" from various related phrases or clauses preceding them

No Comments

No Known Activations