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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.07

2:0.08

3:0.08

4:0.07

5:0.09

6:0.09

7:0.08

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

arium

-1.98

Hunt

-1.91

aire

-1.82

icist

-1.75

arie

-1.72

License

-1.70

ilus

-1.67

Acknowled

-1.67

gat

-1.65

scill

-1.63

POSITIVE LOGITS

 slang

1.87

ipedia

1.70

 amen

1.68

 flashback

1.66

 downside

1.60

Yog

1.58

 calendars

1.56

 Azerbaijan

1.55

aturday

1.53

tru

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.