INDEX

Explanations

phrases related to computer networks and network operations

oai_token-act-pair · gpt-3.5-turbo

references to a "network" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mond

-0.69

ŃĶ

-0.69

orical

-0.67

mar

-0.66

Child

-0.66

ruciating

-0.65

gg

-0.65

vation

-0.65

 Lama

-0.64

ctive

-0.63

POSITIVE LOGITS

 networks

1.07

 network

1.06

 connectivity

1.00

 networking

0.91

 Networks

0.89

 routers

0.86

 footprint

0.84

 connection

0.83

 architectures

0.81

 connections

0.81

Activations Density 0.021%

phrases related to computer networks and network operations

references to a "network" in various contexts

No Comments

No Known Activations