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

1:0.08

2:0.08

3:0.09

4:0.09

5:0.08

6:0.08

7:0.07

8:0.09

9:0.07

10:0.07

11:0.07

Negative Logits

controlled

-2.13

soType

-2.06

Reviewed

-2.00

isine

-1.95

NYSE

-1.92

ECD

-1.90

Initialized

-1.85

stroke

-1.83

Label

-1.82

Software

-1.80

POSITIVE LOGITS

 cove

1.96

 morbid

1.69

 appet

1.63

 includ

1.62

 fancy

1.57

 wonder

1.56

 nightmares

1.56

 trek

1.52

 slack

1.52

 beach

1.50

Activations Density 0.000%

No Known Activations

This feature has no known activations.