INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 behavi

-0.75

 cort

-0.71

 concerned

-0.66

 antioxid

-0.64

 opio

-0.63

 resc

-0.61

 largeDownload

-0.60

 tendon

-0.59

 enthusi

-0.59

displayText

-0.58

POSITIVE LOGITS

lished

0.78

matical

0.75

yip

0.72

number

0.71

matically

0.69

proof

0.69

ritch

0.69

ictionary

0.69

hide

0.68

iland

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.