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

3:0.06

4:0.09

5:0.08

6:0.07

7:0.09

8:0.08

9:0.08

10:0.07

11:0.09

Negative Logits

ovie

-1.81

 Slaughter

-1.73

 exhib

-1.60

 Walls

-1.59

ocaly

-1.55

yrinth

-1.52

allery

-1.51

 camps

-1.50

apse

-1.49

Nation

-1.49

POSITIVE LOGITS

nings

1.94

rules

1.74

 corrected

1.54

ptoms

1.51

 harsher

1.51

 restraining

1.49

 vouchers

1.48

 Balanced

1.45

 Immunity

1.45

 forgiven

1.43

Activations Density 0.000%

No Known Activations

This feature has no known activations.