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

1:0.04

2:0.09

3:0.08

4:0.08

5:0.08

6:0.07

7:0.08

8:0.08

9:0.09

10:0.09

11:0.08

Negative Logits

 outweigh

-1.51

utra

-1.51

iphate

-1.47

 Skydragon

-1.44

 products

-1.44

aters

-1.43

 outwe

-1.42

 Parade

-1.42

 Occupations

-1.40

iner

-1.39

POSITIVE LOGITS

DPR

1.79

stretched

1.58

imately

1.52

rm

1.52

 pending

1.43

DM

1.41

 sanity

1.40

minded

1.39

likely

1.38

confirmed

1.37

Activations Density 0.000%

No Known Activations

This feature has no known activations.