INDEX

Explanations

website or newsletter subscription issues

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.04

1:0.02

2:0.27

3:0.03

4:0.09

5:0.05

6:0.02

7:0.01

8:0.30

9:0.06

10:0.05

11:0.02

Negative Logits

iveness

-1.29

ilies

-1.22

 withhold

-1.18

ammu

-1.17

stood

-1.15

 looming

-1.15

amac

-1.14

ise

-1.11

 celebrates

-1.09

 casts

-1.09

POSITIVE LOGITS

�

1.72

 unnoticed

1.61

Downloadha

1.44

abus

1.38

 overboard

1.36

�

1.35

oslav

1.29

leep

1.25

 interstate

1.22

utical

1.17

Activations Density 0.043%

website or newsletter subscription issues

No Comments

No Known Activations