INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

3:0.06

4:0.07

5:0.08

6:0.08

7:0.07

8:0.08

9:0.09

10:0.07

11:0.09

Negative Logits

onto

-1.39

Course

-1.31

ogly

-1.31

Cal

-1.27

pez

-1.26

anto

-1.25

comfort

-1.25

toc

-1.24

osc

-1.24

Sea

-1.24

POSITIVE LOGITS

══

1.40

 Hindus

1.34

 understatement

1.28

 Khan

1.27

 Criminal

1.27

 bishops

1.26

lance

1.25

��

1.24

 slander

1.23

 meantime

1.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.