INDEX

Explanations

phrases related to legality and governmental 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

1:0.02

2:0.12

3:0.09

4:0.43

5:0.03

6:0.04

7:0.03

8:0.05

9:0.05

10:0.04

11:0.04

Negative Logits

anship

-2.06

aunder

-1.84

ACY

-1.72

auga

-1.62

vP

-1.62

antis

-1.61

aundering

-1.60

OPA

-1.58

iltration

-1.55

urring

-1.55

POSITIVE LOGITS

 maps

1.67

 passports

1.51

 loads

1.51

 beams

1.49

 extracts

1.45

 dictionary

1.44

 receipts

1.42

rams

1.40

 delegates

1.40

 strings

1.40

Activations Density 0.006%

phrases related to legality and governmental actions

No Comments

No Known Activations