INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

1:0.12

2:0.08

3:0.09

4:0.07

5:0.06

6:0.08

7:0.08

8:0.08

9:0.08

10:0.08

11:0.07

Negative Logits

 Thro

-1.66

 blocker

-1.62

oscope

-1.58

Choose

-1.57

ipers

-1.54

 Grammy

-1.53

othal

-1.49

Scroll

-1.46

oscopic

-1.43

 Billboard

-1.42

POSITIVE LOGITS

 Huss

1.80

 Aram

1.59

 appell

1.58

lords

1.56

oglu

1.56

soType

1.54

 disg

1.53

 Maced

1.52

ş

1.50

 Denis

1.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.