INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atche

-0.87

OIL

-0.79

ickets

-0.76

WATCH

-0.69

 pestic

-0.68

CLASSIFIED

-0.65

 Rohing

-0.65

OOL

-0.64

 compost

-0.64

INS

-0.64

POSITIVE LOGITS

op

0.85

conom

0.81

 Presence

0.74

furt

0.73

mma

0.69

nect

0.68

ndum

0.67

 Creed

0.65

phas

0.65

 Ability

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.