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

archive

-0.77

scale

-0.74

arte

-0.72

Clock

-0.72

edged

-0.70

pect

-0.69

orders

-0.69

 Variant

-0.66

UTC

-0.65

CEPT

-0.65

POSITIVE LOGITS

 Gaal

0.68

 exting

0.67

 retention

0.67

éļ

0.65

 Mutual

0.63

 dece

0.62

ãĥ¼ãĥĨ

0.62

 benef

0.62

VERTISEMENT

0.61

ngth

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.