INDEX

Explanations

expressions of skepticism or doubt regarding statements or claims

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

1:0.01

2:0.04

3:0.15

4:0.02

5:0.04

6:0.02

7:0.05

8:0.02

9:0.02

10:0.50

11:0.03

Negative Logits

 during

-1.87

 comparable

-1.84

 dormant

-1.84

 BEFORE

-1.82

 similar

-1.80

 interactive

-1.78

 livestream

-1.75

 while

-1.66

 AFTER

-1.64

 moratorium

-1.63

POSITIVE LOGITS

 understatement

2.31

 guessed

2.21

umph

2.18

haps

2.10

Pwr

2.10

)}

2.06

 exaggeration

2.01

icion

1.96

})

1.95

 guess

1.92

Activations Density 0.454%

expressions of skepticism or doubt regarding statements or claims

No Comments

No Known Activations