INDEX

Explanations

No Explanations Found

New Auto-Interp

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

1:0.07

2:0.08

3:0.08

4:0.08

5:0.08

6:0.08

7:0.07

8:0.09

9:0.07

10:0.09

11:0.09

Negative Logits

 interaction

-1.72

enez

-1.71

 chars

-1.61

 Levy

-1.61

 Philly

-1.60

��

-1.56

 exceptions

-1.56

 Chomsky

-1.55

 Giuliani

-1.54

ゴン

-1.54

POSITIVE LOGITS

Guide

1.82

conservative

1.79

luster

1.75

YP

1.70

Untitled

1.67

Vert

1.63

conserv

1.62

resents

1.60

Ranked

1.59

Rot

1.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.