INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

0:0.08

1:0.08

2:0.07

3:0.08

4:0.09

5:0.09

6:0.07

7:0.08

8:0.07

9:0.08

10:0.09

11:0.07

Negative Logits

 Canadians

-2.69

CAM

-2.53

 registers

-2.47

 franc

-2.47

PCR

-2.47

Cath

-2.43

®,

-2.42

IOC

-2.41

Cam

-2.37

CBC

-2.37

POSITIVE LOGITS

agascar

2.83

Moj

2.73

psc

2.72

Downloadha

2.71

irrel

2.67

 botched

2.66

mares

2.59

██

2.58

wreck

2.52

gross

2.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.