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

0:0.10

1:0.06

2:0.10

3:0.09

4:0.07

5:0.06

6:0.08

7:0.06

8:0.09

9:0.07

10:0.09

11:0.08

Negative Logits

Sources

-1.80

Reference

-1.79

Adv

-1.67

hw

-1.66

media

-1.61

PDATE

-1.61

Media

-1.61

 advoc

-1.60

ETF

-1.56

Weight

-1.55

POSITIVE LOGITS

 shitty

1.79

 crap

1.75

 porous

1.60

 shit

1.59

 blah

1.55

 bullshit

1.55

 nonexistent

1.55

 earthquakes

1.54

 bubbles

1.54

 asshole

1.50

Activations Density 0.000%

No Known Activations

This feature has no known activations.