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

0:0.08

1:0.08

2:0.07

3:0.08

4:0.08

5:0.07

6:0.07

7:0.09

8:0.09

9:0.07

10:0.07

11:0.08

Negative Logits

 behavi

-1.91

aunders

-1.84

 Croat

-1.80

ğ

-1.79

acas

-1.79

ccording

-1.74

rul

-1.73

 behav

-1.71

utsche

-1.70

 disadvant

-1.68

POSITIVE LOGITS

storm

1.77

 Avalon

1.72

ucket

1.59

 Sands

1.59

woods

1.56

 uncovered

1.56

Storm

1.56

 Storm

1.56

Wind

1.56

wind

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.