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

0:0.08

1:0.07

2:0.09

3:0.08

4:0.08

5:0.09

6:0.07

7:0.07

8:0.08

9:0.08

10:0.08

11:0.06

Negative Logits

raq

-1.59

 Sect

-1.52

 thereafter

-1.50

 Immunity

-1.26

 confer

-1.26

 Assist

-1.22

Ped

-1.21

Dep

-1.18

rio

-1.13

REDACTED

-1.13

POSITIVE LOGITS

ktop

1.51

 behavi

1.51

swer

1.50

��

1.45

umers

1.45

 Yelp

1.44

 Calories

1.42

krit

1.33

 blat

1.33

jobs

1.32

Activations Density 0.000%

No Known Activations

This feature has no known activations.