INDEX

Explanations

words that indicate a benefit or positive outcome in a context

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.07

1:0.08

2:0.08

3:0.08

4:0.08

5:0.09

6:0.07

7:0.07

8:0.08

9:0.08

10:0.09

11:0.07

Negative Logits

 beetles

-3.27

 beetle

-3.26

 phantom

-2.76

 eaves

-2.73

 bacter

-2.71

 Wraith

-2.69

 holog

-2.66

 horm

-2.64

 diseases

-2.62

 spores

-2.57

POSITIVE LOGITS

ICO

2.82

 Catalyst

2.81

 Morrison

2.78

ativity

2.65

 Filip

2.62

 Rivals

2.58

 Mana

2.56

 corrid

2.55

 LGBTQ

2.51

 Moreno

2.50

Activations Density 0.000%

words that indicate a benefit or positive outcome in a context

No Known Activations