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

thumbnails

-0.95

agher

-0.78

éĹĺ

-0.75

ividual

-0.73

ngth

-0.72

ãĤ´

-0.69

chemical

-0.68

ãģı

-0.67

duc

-0.64

posts

-0.64

POSITIVE LOGITS

 breath

0.72

ombo

0.69

omb

0.66

op

0.62

op

0.62

opia

0.62

uba

0.61

osaurus

0.61

 Myers

0.61

 recount

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.