INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 concess

-0.75

 traged

-0.72

aukee

-0.71

 attendant

-0.71

 reluct

-0.69

 coerc

-0.67

idth

-0.66

 stall

-0.65

gypt

-0.64

ashtra

-0.63

POSITIVE LOGITS

sheet

0.70

kies

0.70

RAFT

0.67

itchie

0.66

beard

0.65

çīĪ

0.64

owned

0.63

 Medline

0.63

iu

0.62

ensen

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.