INDEX

Explanations

names of individuals or characters and their related activities or roles

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

éĹ

-0.75

Writer

-0.63

 Trave

-0.60

ishable

-0.59

igrants

-0.59

Guide

-0.58

=-=-=-=-=-=-=-=-

-0.58

raved

-0.58

Written

-0.58

olson

-0.57

POSITIVE LOGITS

 replaced

0.65

adj

0.64

 incidentally

0.64

ership

0.63

 herself

0.62

 Democr

0.61

 notoriously

0.60

 overpower

0.59

,,,,

0.59

 receptive

0.58

Activations Density 1.104%

names of individuals or characters and their related activities or roles

No Comments

No Known Activations