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

2:0.07

3:0.07

4:0.08

5:0.08

6:0.07

7:0.08

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

ouble

-1.84

 acceler

-1.64

MpServer

-1.62

Geh

-1.59

):

-1.58

 accelerator

-1.55

profits

-1.53

urga

-1.51

 Canaver

-1.51

 largeDownload

-1.50

POSITIVE LOGITS

1.68

 latitude

1.66

NAS

1.59

GAM

1.59

 Huntington

1.57

 Goodman

1.55

 Dual

1.53

 Rapp

1.52

 Bachelor

1.50

 presumption

1.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.