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

0:0.08

1:0.07

2:0.08

3:0.08

4:0.08

5:0.07

6:0.07

7:0.09

8:0.07

9:0.09

10:0.09

11:0.07

Negative Logits

 Summers

-1.68

 Pearce

-1.60

 Charlottesville

-1.57

 Lerner

-1.55

 Bennett

-1.48

 Maxwell

-1.46

 McDonnell

-1.44

 Conway

-1.43

 Watt

-1.42

 Kessler

-1.41

POSITIVE LOGITS

Reviewer

1.89

qus

1.62

earch

1.60

reb

1.55

soever

1.55

iologist

1.51

vier

1.50

toggle

1.50

vered

1.42

click

1.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.