INDEX

Explanations

phrases related to technology, projects, and events

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

phrases indicating uncertainty or unexpected situations

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

0:0.07

1:0.03

2:0.12

3:0.17

4:0.04

5:0.06

6:0.06

7:0.03

8:0.17

9:0.11

10:0.06

11:0.03

Negative Logits

atives

-1.33

 ItemLevel

-1.26

lication

-1.22

ipers

-1.19

orters

-1.17

notations

-1.17

ahime

-1.14

 monuments

-1.13

boats

-1.13

��

-1.12

POSITIVE LOGITS

 swayed

1.33

awaru

1.28

 coincidence

1.25

sylv

1.18

 interven

1.13

ebook

1.06

Kag

1.05

ogical

1.04

!]

1.03

 bumped

1.02

Activations Density 0.001%

phrases related to technology, projects, and events

phrases indicating uncertainty or unexpected situations

No Comments

No Known Activations