INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 fuelled

-0.68

 thirst

-0.68

»Ĵ

-0.65

entimes

-0.64

 pristine

-0.63

Stre

-0.62

 repaired

-0.62

essel

-0.62

 Patriarch

-0.61

 Sins

-0.60

POSITIVE LOGITS

chini

0.77

dfx

0.77

gins

0.74

umsy

0.69

dress

0.68

uana

0.66

asta

0.66

opa

0.65

olitan

0.64

roxy

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.