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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.07

2:0.06

3:0.10

4:0.10

5:0.07

6:0.11

7:0.07

8:0.06

9:0.09

10:0.05

11:0.06

Negative Logits

 Barnett

-2.59

 Parsons

-2.58

 Barton

-2.53

Shape

-2.36

 Grad

-2.35

Ott

-2.34

-2.31

OW

-2.28

anton

-2.27

 Gust

-2.27

POSITIVE LOGITS

TPP

3.02

 Cthulhu

2.98

ulhu

2.94

 noodles

2.82

olesterol

2.74

 Tibetan

2.71

 Frieza

2.66

 Tibet

2.65

jriwal

2.62

iotics

2.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.