INDEX

Explanations

punctuation marks and conjunctions indicating connectivity between ideas or clauses

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

2:0.09

3:0.05

4:0.42

5:0.02

6:0.06

7:0.12

8:0.02

9:0.03

10:0.05

11:0.07

Negative Logits

 counterparts

-1.81

 contemporaries

-1.72

senal

-1.56

 Pear

-1.48

 colleagues

-1.47

 partners

-1.44

ryu

-1.43

 millenn

-1.42

 Gaia

-1.41

 guides

-1.38

POSITIVE LOGITS

unless

1.69

:(

1.68

emphasis

1.67

"]=>

1.64

heading

1.62

 peacefully

1.60

idation

1.52

APTER

1.50

ication

1.46

killing

1.44

Activations Density 0.012%

punctuation marks and conjunctions indicating connectivity between ideas or clauses

No Comments

No Known Activations