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

1:0.05

2:0.08

3:0.09

4:0.09

5:0.08

6:0.08

7:0.07

8:0.09

9:0.07

10:0.08

11:0.08

Negative Logits

jew

-1.77

Orn

-1.75

ivist

-1.66

oeuv

-1.59

inav

-1.54

cloth

-1.51

 Dance

-1.50

utch

-1.49

 Writ

-1.49

ovan

-1.46

POSITIVE LOGITS

 tremend

1.78

 shitty

1.67

ITS

1.61

 enthusi

1.60

 internally

1.56

 favorable

1.56

 stabilized

1.55

 coordinates

1.55

 favourable

1.54

 stability

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.