INDEX

Explanations

negations and indicators of disbelief or doubt

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

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

0:0.04

1:0.03

2:0.08

3:0.09

4:0.06

5:0.05

6:0.27

7:0.02

8:0.15

9:0.06

10:0.06

11:0.04

Negative Logits

 sparing

-1.70

 narrowly

-1.49

 rotating

-1.39

pleting

-1.32

igning

-1.28

achus

-1.27

 centrally

-1.26

paced

-1.26

 plent

-1.25

 deepening

-1.24

POSITIVE LOGITS

sic

2.29

�

1.88

◼

1.82

�

1.75

�

1.74

LCS

1.58

�

1.55

 …"

1.53

Tes

1.51

�

1.51

Activations Density 0.122%

negations and indicators of disbelief or doubt

No Comments

No Known Activations