INDEX

Explanations

terms related to technology and innovation

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

negative sentiment towards political and social issues

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_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.3.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

0:0.02

1:0.02

2:0.10

3:0.27

4:0.02

5:0.02

6:0.08

7:0.04

8:0.03

9:0.10

10:0.08

11:0.16

Negative Logits

 impulse

-1.21

 specificity

-1.17

 importance

-1.14

 caliber

-1.09

 millisec

-1.07

 urgency

-1.07

 credentials

-1.04

 impression

-1.03

 whim

-1.03

 relevance

-1.02

POSITIVE LOGITS

lished

1.48

furt

1.41

 withd

1.40

perse

1.36

ンジ

1.25

orously

1.24

rency

1.24

ortium

1.23

ocated

1.18

riel

1.18

Activations Density 0.013%

terms related to technology and innovation

negative sentiment towards political and social issues

No Comments

No Known Activations