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

 elim

-0.69

 tremend

-0.67

urion

-0.66

ismo

-0.65

 realism

-0.64

cffff

-0.64

srf

-0.63

oS

-0.63

 karma

-0.61

ãĥĥãĥĪ

-0.61

POSITIVE LOGITS

nyder

0.68

endant

0.68

itatively

0.66

ienced

0.64

 Malt

0.64

enhagen

0.64

places

0.62

stown

0.61

orescent

0.61

 Hearth

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.