INDEX

Explanations

instructions and prompts related to social media engagement and content sharing

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

0:0.02

1:0.01

2:0.09

3:0.31

4:0.11

5:0.03

6:0.05

7:0.08

8:0.04

9:0.05

10:0.09

11:0.06

Negative Logits

ylan

-1.72

emale

-1.65

ev

-1.58

otypes

-1.58

terday

-1.57

boys

-1.47

anes

-1.47

 purportedly

-1.44

 unamb

-1.42

ithe

-1.42

POSITIVE LOGITS

 Thumbnails

1.92

 yourself

1.69

yip

1.69

 periodically

1.67

 wherever

1.65

 🙂

1.62

 bookmark

1.58

Interstitial

1.57

闘

1.57

[|

1.51

Activations Density 0.044%

instructions and prompts related to social media engagement and content sharing

No Comments

No Known Activations