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

0:0.09

1:0.06

2:0.09

3:0.08

4:0.09

5:0.06

6:0.08

7:0.08

8:0.08

9:0.08

10:0.09

11:0.08

Negative Logits

 Sanford

-1.66

 Cary

-1.58

ガ

-1.51

 Robinson

-1.48

Io

-1.47

チ

-1.47

 Juno

-1.46

 Basil

-1.45

 resemblance

-1.44

akia

-1.42

POSITIVE LOGITS

iership

1.93

aundering

1.76

ioxide

1.73

anooga

1.70

igion

1.67

enei

1.63

psey

1.63

reon

1.63

Untitled

1.63

gebra

1.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.