INDEX

Explanations

references to the television series "Dragon Ball Super" and discussions about its episodes

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.11

1:0.04

2:0.08

3:0.11

4:0.05

5:0.04

6:0.08

7:0.18

8:0.04

9:0.03

10:0.05

11:0.13

Negative Logits

 metro

-2.21

McA

-2.21

fax

-2.16

etsk

-2.15

aspers

-2.13

mart

-2.04

minist

-2.02

 Hyundai

-1.98

 Clarks

-1.97

POSITIVE LOGITS

Episode

4.95

 Episode

4.57

 episodes

3.69

 episode

3.64

episode

3.42

isodes

2.83

 Podcast

2.70

Ep

2.69

odcast

2.65

itcher

2.63

Activations Density 0.056%

references to the television series "Dragon Ball Super" and discussions about its episodes

No Comments

No Known Activations