INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

apter

-0.65

ipel

-0.62

 similarities

-0.61

endi

-0.61

 colleagues

-0.61

 Hung

-0.60

Desk

-0.59

 processes

-0.59

âĶĢâĶĢ

-0.58

iability

-0.57

POSITIVE LOGITS

roleum

0.81

dding

0.79

pper

0.79

stice

0.69

senal

0.68

imester

0.68

fu

0.66

security

0.66

istani

0.65

inventoryQuantity

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.