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

0:0.10

1:0.06

2:0.08

3:0.07

4:0.08

5:0.08

6:0.09

7:0.08

8:0.07

9:0.07

10:0.08

11:0.07

Negative Logits

 Enlight

-1.74

htt

-1.73

 indo

-1.69

adr

-1.64

 fuse

-1.64

enh

-1.61

"""

-1.59

 Builder

-1.53

 Frem

-1.53

dh

-1.52

POSITIVE LOGITS

yss

1.90

pmwiki

1.85

watching

1.82

 prey

1.74

MpServer

1.64

ricular

1.59

 irregularities

1.57

 budgets

1.55

DAY

1.54

tro

1.54

Activations Density 0.000%

No Known Activations

This feature has no known activations.