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

0:0.09

1:0.09

2:0.07

3:0.07

4:0.08

5:0.07

6:0.08

7:0.07

8:0.08

9:0.09

10:0.08

11:0.07

Negative Logits

版

-3.08

�

-3.00

hate

-2.96

focused

-2.94

venant

-2.93

verb

-2.84

geist

-2.82

grim

-2.81

GB

-2.78

�

-2.73

POSITIVE LOGITS

 Petroleum

3.36

 petroleum

3.32

Maz

3.02

Haz

2.97

 Railroad

2.96

 Hats

2.91

 marble

2.88

 aeros

2.88

roleum

2.86

Cam

2.79

Activations Density 0.000%

No Known Activations

This feature has no known activations.