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

Head Attr Weights

0:0.09

1:0.09

2:0.10

3:0.07

4:0.06

5:0.08

6:0.08

7:0.07

8:0.07

9:0.08

10:0.07

11:0.08

Negative Logits

},{"

-1.66

ネ

-1.51

sonian

-1.49

SET

-1.45

VAL

-1.43

Wil

-1.41

POL

-1.40

ivot

-1.40

 recomm

-1.39

Correct

-1.36

POSITIVE LOGITS

 encount

1.75

acco

1.71

 enthusi

1.66

mort

1.48

 snipers

1.47

ickets

1.46

 veter

1.46

ants

1.42

girls

1.41

 commercials

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.