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

1:0.09

2:0.08

3:0.07

4:0.08

5:0.08

6:0.08

7:0.08

8:0.08

9:0.06

10:0.09

11:0.07

Negative Logits

 Eater

-1.93

 Communications

-1.76

 burg

-1.74

 annoy

-1.64

dale

-1.64

 Trojan

-1.64

 Galaxy

-1.63

 gossip

-1.63

 upset

-1.60

 crow

-1.59

POSITIVE LOGITS

ebted

2.45

=~=~

2.06

fig

1.93

FIG

1.90

PF

1.86

imaru

1.81

[+

1.81

itled

1.80

lethal

1.78

Ply

1.78

Activations Density 0.000%

No Known Activations

This feature has no known activations.