INDEX

Explanations

interactive elements and calls to action within the text

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

2:0.16

3:0.16

4:0.06

5:0.02

6:0.09

7:0.07

8:0.08

9:0.04

10:0.16

11:0.09

Negative Logits

 noun

-1.59

 appell

-1.45

 distilled

-1.43

 cond

-1.42

 offic

-1.41

 hurd

-1.36

 Offic

-1.35

 bragging

-1.34

 Prec

-1.33

 Plains

-1.30

POSITIVE LOGITS

aido

1.61

 Nightmares

1.57

clips

1.55

"}

1.54

Video

1.48

�

1.46

phabet

1.45

swer

1.44

=#

1.44

Leary

1.42

Activations Density 0.001%

interactive elements and calls to action within the text

No Comments

No Known Activations