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

1:0.05

2:0.08

3:0.10

4:0.09

5:0.05

6:0.07

7:0.10

8:0.06

9:0.07

10:0.10

11:0.11

Negative Logits

Management

-1.67

,''

-1.62

 electing

-1.60

],"

-1.57

 administr

-1.52

),"

-1.51

 nationally

-1.45

 migrating

-1.44

''.

-1.43

 renov

-1.42

POSITIVE LOGITS

ニ

1.59

itsu

1.54

ヘ

1.53

ça

1.51

1.50

adesh

1.47

agnetic

1.45

olen

1.44

jab

1.44

isu

1.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.