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

1:0.06

2:0.09

3:0.08

4:0.09

5:0.07

6:0.07

7:0.08

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

Reviewer

-2.07

────

-1.80

lex

-1.70

 compilation

-1.67

oples

-1.67

OLOG

-1.63

 stru

-1.62

 explan

-1.61

rien

-1.59

 organising

-1.57

POSITIVE LOGITS

 gunshots

1.74

 Missile

1.66

arms

1.66

ミ

1.62

bright

1.62

robe

1.60

AIDS

1.59

 gunfire

1.57

anamo

1.56

 theaters

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.