INDEX

Explanations

instances of attributive phrases and references

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.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

0:0.01

1:0.04

2:0.21

3:0.14

4:0.02

5:0.03

6:0.06

7:0.06

8:0.05

9:0.14

10:0.09

11:0.10

Negative Logits

assies

-1.44

apo

-1.27

anamo

-1.24

ivot

-1.19

ierce

-1.15

waukee

-1.15

ARI

-1.13

ruce

-1.12

ojure

-1.12

termin

-1.12

POSITIVE LOGITS

 hindsight

1.11

 circumstances

1.11

 deed

1.10

 conscience

1.05

 compass

1.04

 ante

1.03

 precept

1.02

 percentage

1.00

 sane

1.00

 hell

1.00

Activations Density 0.036%

instances of attributive phrases and references

No Comments

No Known Activations