INDEX

Explanations

phrases expressing uncertainty or questioning decisions

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

0:0.02

1:0.02

2:0.18

3:0.08

4:0.10

5:0.02

6:0.02

7:0.32

8:0.04

9:0.03

10:0.06

11:0.06

Negative Logits

untled

-1.98

 reputable

-1.85

 trustworthy

-1.80

authorized

-1.76

uyomi

-1.74

quickShipAvailable

-1.73

igham

-1.70

emark

-1.63

 independ

-1.55

soType

-1.54

POSITIVE LOGITS

 Caps

1.60

 Soccer

1.54

ilda

1.53

PCR

1.52

RAW

1.51

WWE

1.47

 Boxing

1.46

 Struggle

1.40

 Brawl

1.38

 Paint

1.36

Activations Density 0.012%

phrases expressing uncertainty or questioning decisions

No Comments

No Known Activations