INDEX

Explanations

No Explanations Found

New Auto-Interp

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

0:0.08

1:0.08

2:0.10

3:0.08

4:0.07

5:0.07

6:0.07

7:0.09

8:0.07

9:0.09

10:0.08

11:0.07

Negative Logits

 therap

-1.72

 desper

-1.70

ixtape

-1.66

 happening

-1.63

 behav

-1.62

 proble

-1.62

 advoc

-1.54

 diaper

-1.54

peat

-1.50

ule

-1.48

POSITIVE LOGITS

USS

1.55

Reply

1.47

�

1.47

 srfAttach

1.44

mon

1.41

 Tsukuyomi

1.39

 Metropolitan

1.37

oted

1.37

Journal

1.36

 Investor

1.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.