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

cules

-0.91

deen

-0.72

toe

-0.67

enegger

-0.67

roman

-0.64

faces

-0.64

alter

-0.64

sheets

-0.63

finger

-0.63

cular

-0.62

POSITIVE LOGITS

~~~~

0.68

ikk

0.64

 incap

0.64

0.62

"]

0.60

 "$:/

0.60

ifest

0.60

å¸

0.59

ï¸

0.59

IENT

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.