INDEX

Explanations

specific structural elements or formatting of 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

0:0.16

1:0.13

2:0.05

3:0.07

4:0.04

5:0.11

6:0.03

7:0.02

8:0.07

9:0.09

10:0.11

11:0.07

Negative Logits

killer

-1.89

iannopoulos

-1.69

\/

-1.65

love

-1.55

ube

-1.51

sex

-1.49

hide

-1.46

opian

-1.44

classes

-1.44

POSITIVE LOGITS

 Municipal

1.64

 Miscellaneous

1.55

 Estimated

1.55

��

1.51

 Instruction

1.49

 Official

1.44

 Contents

1.43

 FedEx

1.43

 OFFIC

1.43

 Illegal

1.42

Activations Density 0.007%

specific structural elements or formatting of text

No Comments

No Known Activations