INDEX

Explanations

instances of punctuation and pauses in 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

0:0.24

1:0.05

2:0.05

3:0.08

4:0.04

5:0.07

6:0.03

7:0.02

8:0.16

9:0.07

10:0.05

11:0.08

Negative Logits

 labeled

-1.73

 referenced

-1.56

 plural

-1.50

 Larson

-1.50

 understatement

-1.48

 labelled

-1.45

 portrayed

-1.45

 Parables

-1.45

 Tuls

-1.44

 Picard

-1.42

POSITIVE LOGITS

vir

1.76

irc

1.70

wich

1.67

imore

1.64

cig

1.61

aven

1.60

oven

1.60

anasia

1.58

rag

1.57

shop

1.55

Activations Density 0.001%

instances of punctuation and pauses in text

No Comments

No Known Activations