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

1:0.10

2:0.08

3:0.08

4:0.08

5:0.08

6:0.06

7:0.07

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

['

-1.73

 Huntington

-1.67

ewater

-1.62

ken

-1.57

unknown

-1.56

 intervening

-1.56

Nation

-1.54

aughlin

-1.53

advertisement

-1.51

mont

-1.49

POSITIVE LOGITS

 SHALL

1.72

oan

1.64

 teenagers

1.46

 cohesive

1.45

YPG

1.45

itiz

1.44

 Grail

1.43

ixtape

1.42

IBLE

1.41

ALE

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.