INDEX

Explanations

phrases and words indicating strength and resilience

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.27

1:0.03

2:0.19

3:0.06

4:0.04

5:0.06

6:0.05

7:0.04

8:0.02

9:0.07

10:0.06

11:0.05

Negative Logits

uberty

-3.00

urnal

-2.64

Narr

-2.51

 Accuracy

-2.48

ellectual

-2.46

elligence

-2.38

 auditory

-2.34

ocrine

-2.33

 Berkshire

-2.32

 Ranking

-2.31

POSITIVE LOGITS

 glue

5.98

 adhesive

4.50

 glued

3.44

 bonding

3.23

 screws

3.21

 stitching

3.08

 solvent

3.08

 binds

2.98

gum

2.91

 sticky

2.85

Activations Density 0.001%

phrases and words indicating strength and resilience

No Comments

No Known Activations