INDEX

Explanations

phrases related to goals and achievements in a crowdfunding context

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_L5_Hcat_z_lr1.20e-03_l11.00e+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.5.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

0:0.02

1:0.02

2:0.06

3:0.09

4:0.05

5:0.05

6:0.03

7:0.04

8:0.04

9:0.06

10:0.34

11:0.14

Negative Logits

 Scully

-1.77

 Beir

-1.70

 Arpaio

-1.60

 Hannity

-1.52

 arresting

-1.41

Tup

-1.37

 Rath

-1.35

 vans

-1.25

 Melania

-1.25

 Scalia

-1.23

POSITIVE LOGITS

eco

1.57

ulkan

1.43

ename

1.42

 increments

1.41

方

1.41

ynes

1.33

buy

1.33

 Desktop

1.33

alted

1.33

 Developer

1.31

Activations Density 0.032%

phrases related to goals and achievements in a crowdfunding context

No Comments

No Known Activations