INDEX

Explanations

negative sentiments or actions related to authority and governance

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.02

2:0.16

3:0.20

4:0.04

5:0.05

6:0.02

7:0.03

8:0.10

9:0.09

10:0.10

11:0.11

Negative Logits

ritten

-1.34

rification

-1.32

atal

-1.29

 pione

-1.22

rament

-1.19

ategor

-1.17

lance

-1.13

bear

-1.12

 Leban

-1.09

 Polar

-1.09

POSITIVE LOGITS

 "...

1.50

 Pastebin

1.38

 "…

1.33

 himself

1.29

 herself

1.29

 tone

1.27

gging

1.27

Dialogue

1.25

Instead

1.24

"[

1.22

Activations Density 0.392%

negative sentiments or actions related to authority and governance

No Comments

No Known Activations