INDEX

Explanations

statistical or numerical information related to different subjects

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

references to significant shifts or changes in societal structures or political landscapes

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

1:0.06

2:0.08

3:0.01

4:0.03

5:0.14

6:0.08

7:0.08

8:0.06

9:0.06

10:0.08

11:0.27

Negative Logits

��

-1.64

verett

-1.63

���

-1.56

))

-1.54

plin

-1.53

 expired

-1.46

else

-1.41

arten

-1.40

 deleted

-1.40

Downloadha

-1.38

POSITIVE LOGITS

 lately

1.88

especially

1.83

particularly

1.82

 Particularly

1.75

 Especially

1.74

 Opposition

1.71

 especially

1.67

Especially

1.63

 nowadays

1.61

ndra

1.58

Activations Density 0.385%

statistical or numerical information related to different subjects

references to significant shifts or changes in societal structures or political landscapes

No Comments

No Known Activations