INDEX

Explanations

mentions of political figures and policies

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

0:0.09

1:0.05

2:0.04

3:0.06

4:0.03

5:0.08

6:0.03

7:0.11

8:0.04

9:0.24

10:0.12

11:0.07

Negative Logits

 gradient

-1.19

 seams

-1.19

 lineback

-1.13

 thickness

-1.11

 approximation

-1.10

 difference

-1.10

 density

-1.10

 errors

-1.09

 grad

-1.09

vag

-1.08

POSITIVE LOGITS

�士

1.66

 veto

1.31

assad

1.26

 appoint

1.22

Executive

1.19

 Himself

1.19

 himself

1.19

reprene

1.14

isan

1.11

 Truman

1.10

Activations Density 0.199%

mentions of political figures and policies

No Comments

No Known Activations