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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.08

2:0.08

3:0.09

4:0.07

5:0.08

6:0.08

7:0.08

8:0.08

9:0.08

10:0.07

11:0.08

Negative Logits

 enthusi

-2.20

eatures

-2.20

 fools

-2.11

 spons

-1.90

 adolesc

-1.85

 reluct

-1.81

 murd

-1.77

 describ

-1.75

 fung

-1.75

 nodd

-1.75

POSITIVE LOGITS

location

2.12

 Unknown

1.94

ubi

1.83

 Activ

1.80

tained

1.78

�

1.75

��

1.73

 Completed

1.72

�

1.70

 Change

1.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.