INDEX

Explanations

phrases related to redefining concepts or characteristics in various 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_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

0:0.01

1:0.01

2:0.05

3:0.07

4:0.19

5:0.02

6:0.02

7:0.37

8:0.03

9:0.03

10:0.08

11:0.07

Negative Logits

enza

-1.56

 terminating

-1.50

eret

-1.46

 periodic

-1.36

 drawback

-1.35

 Warning

-1.34

 upkeep

-1.33

 redund

-1.32

 Fenrir

-1.32

isy

-1.31

POSITIVE LOGITS

 boundaries

1.78

 norms

1.61

terday

1.54

yrights

1.54

onomy

1.54

pires

1.50

 redef

1.47

rules

1.46

 parad

1.46

kens

1.44

Activations Density 0.002%

phrases related to redefining concepts or characteristics in various contexts

No Comments

No Known Activations