INDEX

Explanations

phrases and terms related to learning and lessons

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.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

2:0.02

3:0.05

4:0.07

5:0.08

6:0.03

7:0.01

8:0.32

9:0.05

10:0.01

11:0.02

Negative Logits

aucuses

-1.74

epad

-1.66

ideos

-1.64

forums

-1.55

ournaments

-1.50

 arsen

-1.48

ittees

-1.47

 cabinets

-1.46

 snakes

-1.45

atoon

-1.42

POSITIVE LOGITS

 wisdom

1.73

 reinvest

1.59

tid

1.57

 McCann

1.56

 Dull

1.55

elsen

1.54

 insight

1.54

 footing

1.40

 Relief

1.39

 DACA

1.35

Activations Density 0.004%

phrases and terms related to learning and lessons

No Comments

No Known Activations