INDEX

Explanations

No Explanations Found

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

0:0.09

1:0.07

2:0.08

3:0.08

4:0.07

5:0.08

6:0.08

7:0.08

8:0.09

9:0.08

10:0.08

11:0.06

Negative Logits

 Cher

-1.86

 Krishna

-1.85

Pinterest

-1.82

Tie

-1.80

 Graveyard

-1.67

 Pinterest

-1.66

Mim

-1.66

Rin

-1.63

 Basil

-1.63

Tid

-1.60

POSITIVE LOGITS

 taxing

1.77

module

1.74

regulated

1.73

volume

1.72

 certific

1.72

mods

1.69

ffect

1.68

 lawfully

1.64

authorized

1.60

iso

1.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.