INDEX

Explanations

attends to code-related elements from generic function-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

0:0.64

1:0.02

2:0.02

3:0.01

4:0.14

5:0.05

6:0.03

7:0.05

Negative Logits

-0.33

<eos>

-0.33

log

-0.32

-0.31

-0.30

-0.29

-0.28

POSITIVE LOGITS

tagext

0.51

RegistryLite

0.49

__':

0.44

abestanden

0.43

CloseOperation

0.43

 initComponents

0.42

featureID

0.41

 nakalista

0.41

complexContent

0.41

 '\\;'

0.40

Activations Density 0.005%

attends to code-related elements from generic function-related tokens

No Comments

No Known Activations