INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.10

2:0.08

3:0.07

4:0.07

5:0.08

6:0.07

7:0.08

8:0.06

9:0.09

10:0.08

11:0.08

Negative Logits

 Juda

-2.28

 perjury

-2.03

 bargain

-1.80

 myster

-1.78

 handcuffs

-1.68

 adjourn

-1.68

 reconc

-1.67

 wrongful

-1.66

 heap

-1.66

 comprom

-1.64

POSITIVE LOGITS

2.26

dra

2.18

�

1.91

OPA

1.90

 guiActiveUn

1.90

fleet

1.90

reb

1.85

 Unloaded

1.82

jam

1.77

ergic

1.76

Activations Density 0.000%

No Known Activations

This feature has no known activations.