INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.09

2:0.07

3:0.08

4:0.08

5:0.07

6:0.07

7:0.08

8:0.08

9:0.10

10:0.09

11:0.07

Negative Logits

Numbers

-3.30

 Hebdo

-2.95

acies

-2.77

 Codex

-2.75

 Constantine

-2.75

ACY

-2.66

 Gaming

-2.63

 Hermes

-2.56

isine

-2.54

 Fact

-2.54

POSITIVE LOGITS

berman

3.14

utor

2.72

 rods

2.63

…………

2.54

 lear

2.52

 encl

2.49

 adjustable

2.48

 marsh

2.46

 petitioner

2.45

ilateral

2.42

Activations Density 0.000%

No Known Activations

This feature has no known activations.