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

0:0.07

1:0.09

2:0.09

3:0.08

4:0.09

5:0.08

6:0.07

7:0.09

8:0.08

9:0.06

10:0.07

11:0.07

Negative Logits

opoly

-2.17

��

-2.04

wra

-2.00

onomy

-1.89

 Mecca

-1.82

abase

-1.74

archy

-1.73

inar

-1.72

 Roll

-1.72

 Regions

-1.72

POSITIVE LOGITS

nz

2.23

NZ

1.80

 premature

1.77

 junior

1.72

tem

1.68

 assault

1.60

 reckless

1.60

hon

1.59

 disagrees

1.58

 suicidal

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.