INDEX

Explanations

words related to politics and government affairs

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

terms related to negative implications and consequences, especially in the context of reputation and moral judgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.06

1:0.03

2:0.43

3:0.06

4:0.11

5:0.05

6:0.01

7:0.02

8:0.06

9:0.05

10:0.05

11:0.02

Negative Logits

uddin

-1.38

 pockets

-1.24

querque

-1.22

umbai

-1.21

 pond

-1.17

 sophistication

-1.17

oha

-1.13

QB

-1.11

ibrary

-1.10

arius

-1.10

POSITIVE LOGITS

ctive

1.50

ction

1.44

lishes

1.42

ciation

1.41

ォ

1.38

lished

1.36

issance

1.32

elected

1.23

untled

1.19

ancing

1.18

Activations Density 0.012%

words related to politics and government affairs

terms related to negative implications and consequences, especially in the context of reputation and moral judgment

No Comments

No Known Activations