INDEX

Explanations

No Explanations Found

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

0:0.08

1:0.05

2:0.09

3:0.09

4:0.09

5:0.08

6:0.07

7:0.07

8:0.08

9:0.08

10:0.08

11:0.09

Negative Logits

 Revised

-1.89

 Signed

-1.87

 Scholars

-1.76

 Commander

-1.73

 Painter

-1.72

named

-1.72

editor

-1.72

Pri

-1.71

GMT

-1.67

 Advocate

-1.66

POSITIVE LOGITS

lections

2.07

ibaba

2.05

usterity

1.96

atility

1.94

osphere

1.93

 outweigh

1.90

 exceeds

1.84

ularity

1.84

ogie

1.81

emouth

1.80

Activations Density 0.000%

No Known Activations

This feature has no known activations.