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

0:0.07

1:0.06

2:0.10

3:0.08

4:0.09

5:0.08

6:0.07

7:0.09

8:0.07

9:0.07

10:0.08

11:0.09

Negative Logits

culosis

-1.69

 Pavel

-1.67

naissance

-1.65

undown

-1.64

 culminated

-1.61

ariat

-1.61

riots

-1.60

ixties

-1.58

aughs

-1.56

 Bruins

-1.54

POSITIVE LOGITS

amazon

1.66

channelAvailability

1.66

cup

1.65

feed

1.62

aph

1.61

 donation

1.60

access

1.58

AppData

1.57

mod

1.57

Untitled

1.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.