INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 manufact

-0.80

 coating

-0.68

 Slay

-0.67

 enthusi

-0.66

 Ames

-0.66

 introducing

-0.65

 condem

-0.63

angelo

-0.62

oving

-0.61

 moisture

-0.61

POSITIVE LOGITS

veyard

0.81

itar

0.73

 Attribution

0.69

riors

0.68

GBT

0.67

{"

0.65

cdn

0.65

wave

0.64

estamp

0.64

 Curve

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.