INDEX

Explanations

terms associated with political power struggles and rights issues, particularly relating to Russia and immigration

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

0:0.20

1:0.05

2:0.02

3:0.07

4:0.10

5:0.08

6:0.06

7:0.14

8:0.03

9:0.06

10:0.03

11:0.11

Negative Logits

theless

-2.67

lished

-2.39

 namely

-2.09

everal

-2.06

ospons

-2.06

 likewise

-2.04

 abundantly

-2.02

 privately

-1.99

 leapt

-1.97

 nour

-1.93

POSITIVE LOGITS

 Tags

2.37

 Taxes

2.26

ICS

2.26

olitics

2.13

 Misc

2.01

ORTS

1.97

 Rape

1.95

!,

1.93

culture

1.91

,...

1.91

Activations Density 0.006%

terms associated with political power struggles and rights issues, particularly relating to Russia and immigration

No Comments

No Known Activations