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

1:0.08

2:0.07

3:0.07

4:0.07

5:0.08

6:0.07

7:0.06

8:0.09

9:0.09

10:0.10

11:0.08

Negative Logits

 react

-1.70

 strain

-1.65

hol

-1.52

illard

-1.48

 dies

-1.46

emia

-1.45

 meanwhile

-1.45

 reck

-1.45

etc

-1.43

 Born

-1.42

POSITIVE LOGITS

DVD

1.86

worldly

1.85

CAST

1.79

UCT

1.65

PATH

1.65

YC

1.64

EO

1.63

ensual

1.62

song

1.56

ginx

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.