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

0:0.06

1:0.10

2:0.07

3:0.08

4:0.08

5:0.08

6:0.09

7:0.08

8:0.07

9:0.08

10:0.08

11:0.07

Negative Logits

 bench

-1.73

CPC

-1.69

GBT

-1.68

aples

-1.67

Vert

-1.65

 plurality

-1.60

realDonaldTrump

-1.59

 Huntington

-1.58

 Oval

-1.57

obl

-1.54

POSITIVE LOGITS

actionDate

2.09

raq

2.05

rul

1.96

̶

1.85

agara

1.81

Sai

1.81

ategory

1.74

rir

1.71

roo

1.68

)</

1.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.