INDEX

Explanations

phrases indicating past experiences or actions

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

0:0.03

1:0.01

2:0.30

3:0.22

4:0.11

5:0.03

6:0.02

7:0.05

8:0.04

9:0.04

10:0.05

11:0.05

Negative Logits

versely

-1.54

 habit

-1.48

tails

-1.47

bp

-1.41

requisites

-1.40

 tails

-1.40

ager

-1.38

versions

-1.38

KL

-1.37

 veto

-1.36

POSITIVE LOGITS

sbm

1.81

ogyn

1.58

secure

1.49

ilipp

1.49

 Mortal

1.48

 Palestin

1.41

 devoted

1.39

Saf

1.37

 corrid

1.36

itals

1.34

Activations Density 0.000%

phrases indicating past experiences or actions

No Comments

No Known Activations