INDEX

Explanations

references to personal or healthcare-related responsibilities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.01

2:0.08

3:0.11

4:0.27

5:0.03

6:0.08

7:0.19

8:0.02

9:0.03

10:0.06

11:0.04

Negative Logits

BuyableInstoreAndOnline

-1.60

untu

-1.57

busters

-1.46

icted

-1.45

pires

-1.45

���

-1.41

poons

-1.41

odder

-1.39

usal

-1.39

annis

-1.38

POSITIVE LOGITS

 detail

1.73

 strengths

1.62

=#

1.62

 heterogeneity

1.60

pillar

1.57

 fronts

1.56

 Remastered

1.54

 aspect

1.49

":"/

1.47

 nuances

1.46

Activations Density 0.004%

references to personal or healthcare-related responsibilities

No Comments

No Known Activations