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

4:0.07

5:0.08

6:0.08

7:0.09

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

iform

-1.95

 strengths

-1.91

 Cosponsors

-1.71

 solicit

-1.70

ypes

-1.69

 insign

-1.69

ibles

-1.67

xual

-1.64

tumblr

-1.58

 incompet

-1.56

POSITIVE LOGITS

ender

1.89

��

1.75

>>>

1.56

jandro

1.50

undo

1.49

DEP

1.47

 Bitcoins

1.45

��

1.45

 Arena

1.45

rawler

1.43

Activations Density 0.000%

No Known Activations

This feature has no known activations.