INDEX

Explanations

references to positions or roles in sports contexts

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_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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.04

1:0.03

2:0.14

3:0.06

4:0.11

5:0.08

6:0.03

7:0.03

8:0.13

9:0.22

10:0.05

11:0.03

Negative Logits

ologne

-1.27

 Janeiro

-1.20

isine

-1.15

00200000

-1.14

\\\\\\\\

-1.13

ouf

-1.10

zbollah

-1.10

zin

-1.09

etheless

-1.09

ensable

-1.09

POSITIVE LOGITS

 flats

1.62

 vacated

1.45

 slots

1.38

 shelves

1.35

Hub

1.27

 desks

1.21

 precon

1.10

 stones

1.10

stones

1.10

 Slot

1.09

Activations Density 0.006%

references to positions or roles in sports contexts

No Comments

No Known Activations