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

unciation

-0.83

oval

-0.67

iture

-0.66

ality

-0.66

 Revision

-0.66

ishing

-0.64

quet

-0.63

iguous

-0.60

rogen

-0.59

clusion

-0.58

POSITIVE LOGITS

cffffcc

0.74

 srfAttach

0.73

 actionGroup

0.70

 meters

0.70

meter

0.67

Kar

0.67

CLASSIFIED

0.67

psey

0.66

 externalToEVAOnly

0.65

kus

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.