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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.05

2:0.07

3:0.08

4:0.09

5:0.08

6:0.07

7:0.07

8:0.09

9:0.08

10:0.09

11:0.09

Negative Logits

 unexpl

-1.69

vered

-1.56

 indist

-1.55

 blister

-1.54

skip

-1.53

unp

-1.50

 hidden

-1.50

 overflowing

-1.50

 unman

-1.48

 undet

-1.48

POSITIVE LOGITS

 Cosponsors

2.01

affle

1.72

ァ

1.71

affles

1.68

reciation

1.67

 Flavoring

1.66

Ori

1.65

Moe

1.61

actionDate

1.61

 constitu

1.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.