INDEX

Explanations

phrases indicating awareness, assessment, and decision-making processes

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_L7_Hcat_z_lr1.20e-03_l11.10e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.7.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

0:0.12

1:0.03

2:0.01

3:0.09

4:0.07

5:0.12

6:0.03

7:0.02

8:0.11

9:0.31

10:0.01

11:0.02

Negative Logits

 unsurprisingly

-1.93

anmar

-1.88

 hypocritical

-1.86

 moreover

-1.82

livious

-1.79

 additionally

-1.74

 hypoc

-1.68

 hypocr

-1.66

 chiefly

-1.65

uably

-1.64

POSITIVE LOGITS

 arrives

2.13

 arrive

2.07

 presses

1.88

 finished

1.86

 exits

1.85

mop

1.83

 arrived

1.81

 clicked

1.80

 enters

1.79

 expires

1.78

Activations Density 0.009%

phrases indicating awareness, assessment, and decision-making processes

No Comments

No Known Activations