INDEX

Explanations

actions and events related to criminal activities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

phrases indicating support and struggle in challenging situations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.06

1:0.04

2:0.05

3:0.11

4:0.03

5:0.07

6:0.09

7:0.11

8:0.04

9:0.10

10:0.17

11:0.09

Negative Logits

 periphery

-1.03

 guiActiveUnfocused

-0.95

 converge

-0.93

mination

-0.89

 backdrop

-0.86

 ripple

-0.86

 discrep

-0.85

 atmosphere

-0.83

 cohesion

-0.82

 differential

-0.82

POSITIVE LOGITS

 himself

1.42

STD

1.02

 herself

1.01

 Himself

0.99

AFTA

0.92

TIT

0.87

Tact

0.86

KN

0.86

TAG

0.83

estern

0.83

Activations Density 0.393%

actions and events related to criminal activities

phrases indicating support and struggle in challenging situations

No Comments

No Known Activations