INDEX

Explanations

phrases related to news articles, government actions, and incidents

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

phrases indicating actions and decisions involving individuals or groups

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

0:0.16

1:0.04

2:0.08

3:0.17

4:0.03

5:0.12

6:0.09

7:0.02

8:0.04

9:0.08

10:0.09

11:0.03

Negative Logits

>.

-1.39

ande

-1.37

%.

-1.34

etheless

-1.32

!".

-1.28

atever

-1.25

$.

-1.22

srf

-1.19

".

-1.18

ptin

-1.17

POSITIVE LOGITS

pires

1.45

)]

1.16

 reproduce

1.12

 photographed

1.03

ottesville

1.00

 filmed

0.98

 surname

0.94

 varies

0.93

?)

0.93

 MISS

0.93

Activations Density 0.163%

phrases related to news articles, government actions, and incidents

phrases indicating actions and decisions involving individuals or groups

No Comments

No Known Activations