INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

3:0.09

4:0.08

5:0.07

6:0.08

7:0.07

8:0.07

9:0.09

10:0.08

11:0.07

Negative Logits

�

-2.94

OTE

-2.67

CPI

-2.65

 taxp

-2.63

lift

-2.61

ulic

-2.60

bucks

-2.60

oppable

-2.57

"}],"

-2.55

throp

-2.54

POSITIVE LOGITS

 Manga

3.05

 Idol

3.05

 Shark

2.88

 Abyss

2.70

 Coral

2.68

 Studio

2.67

 motif

2.56

 Document

2.55

 Atlantis

2.55

 Costume

2.54

Activations Density 0.000%

No Known Activations

This feature has no known activations.