INDEX

Explanations

medical and scientific terminology related to drug administration and effects

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.11

1:0.06

2:0.09

3:0.12

4:0.08

5:0.07

6:0.08

7:0.08

8:0.05

9:0.06

10:0.07

11:0.07

Negative Logits

isSpecialOrderable

-2.57

✓

-2.34

 Medieval

-2.33

「

-2.30

アル

-2.26

�

-2.23

ku

-2.23

 Affordable

-2.23

 Popular

-2.20

�

-2.20

POSITIVE LOGITS

 dopamine

3.67

aline

3.32

 saline

3.25

THC

2.99

 MDMA

2.99

 cocaine

2.99

Amp

2.96

 cannabinoid

2.90

entanyl

2.80

 overnight

2.79

Activations Density 0.011%

medical and scientific terminology related to drug administration and effects

No Known Activations