INDEX

Explanations

attends to version numbers from numeric tokens or references that indicate a version

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

1:0.03

2:0.04

3:0.02

4:0.05

5:0.04

6:0.03

7:0.72

Negative Logits

 myſelf

-1.69

 itſelf

-1.65

Efq

-1.60

 Monfieur

-1.56

 themſelves

-1.50

 Theſe

-1.49

 Anſ

-1.48

 Houſe

-1.46

 Diſ

-1.45

 houſe

-1.45

POSITIVE LOGITS

0.79

0.77

0.71

0.68

<eos>

0.67

“

0.66

0.65

0.59

Activations Density 0.090%

attends to version numbers from numeric tokens or references that indicate a version

No Comments

No Known Activations