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

0:0.07

1:0.07

2:0.09

3:0.08

4:0.08

5:0.09

6:0.06

7:0.08

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

Tickets

-1.83

 Tickets

-1.71

 Mechdragon

-1.66

abi

-1.60

edition

-1.50

Mub

-1.47

phy

-1.46

Rated

-1.45

spons

-1.41

 Audio

-1.40

POSITIVE LOGITS

ocent

1.93

ajor

1.73

ャ

1.67

アル

1.65

osaurus

1.64

gom

1.60

aeda

1.56

├

1.54

.?

1.52

 prow

1.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.