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

adobe

-0.69

 Uriel

-0.66

²¾

-0.66

 Angola

-0.65

Ļ½

-0.63

onent

-0.62

>:

-0.62

igslist

-0.61

PAC

-0.60

(%

-0.60

POSITIVE LOGITS

fried

0.67

paper

0.66

 craw

0.64

glers

0.64

 guesses

0.63

ollow

0.62

breaking

0.59

 prank

0.59

translation

0.58

 Bangl

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.