INDEX

Explanations

references to single entities or items, often in a context of exclusivity or uniqueness

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.05

1:0.06

2:0.07

3:0.07

4:0.08

5:0.04

6:0.06

7:0.08

8:0.03

9:0.15

10:0.04

11:0.22

Negative Logits

 actions

-2.11

inosaur

-2.06

 action

-2.02

roman

-2.02

sth

-1.99

epad

-1.97

 gobl

-1.90

 edits

-1.84

ide

-1.76

written

-1.75

POSITIVE LOGITS

 nickname

2.36

 moniker

2.32

 acronym

2.19

 Grade

2.08

 nick

2.08

 pseudonym

2.02

 initials

1.97

enders

1.95

 calib

1.81

 surn

1.80

Activations Density 0.005%

references to single entities or items, often in a context of exclusivity or uniqueness

No Comments

No Known Activations