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

0:0.09

1:0.07

2:0.08

3:0.08

4:0.07

5:0.09

6:0.08

7:0.08

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

 Scrolls

-1.94

 Spaces

-1.93

 Reno

-1.88

UFO

-1.85

 Krypt

-1.80

 Probe

-1.75

Nab

-1.73

Sph

-1.73

 Verse

-1.71

 UFOs

-1.71

POSITIVE LOGITS

mble

2.66

ependent

2.17

uple

2.05

 withd

1.93

quished

1.92

legates

1.92

ワン

1.87

achelor

1.78

ctors

1.75

 stret

1.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.