INDEX

Explanations

phrases indicating communication or dialogue

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

1:0.04

2:0.07

3:0.06

4:0.17

5:0.04

6:0.04

7:0.26

8:0.03

9:0.04

10:0.09

11:0.08

Negative Logits

enegger

-1.29

 gamb

-1.28

 Applicant

-1.27

 defe

-1.26

ansion

-1.25

 vested

-1.25

thought

-1.24

 attachment

-1.24

 Formation

-1.23

ructure

-1.23

POSITIVE LOGITS

 hourly

1.52

Dispatch

1.48

ophon

1.47

Elsewhere

1.47

 alerts

1.45

 sightings

1.43

 anecd

1.41

 Reports

1.39

 opin

1.38

oday

1.38

Activations Density 0.000%

phrases indicating communication or dialogue

No Comments

No Known Activations