INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Quote

-0.85

 quote

-0.68

 BIOS

-0.64

ollah

-0.63

}}}

-0.61

INAL

-0.61

buster

-0.61

Gas

-0.60

CSS

-0.60

IPM

-0.59

POSITIVE LOGITS

haar

0.81

ylum

0.77

stellar

0.71

istries

0.68

 Primal

0.66

pherd

0.64

loo

0.63

ifacts

0.61

adata

0.60

uthor

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.