Jon Watte

When I want to bind to all interfaces, and specify host='0.0.0.0' in butterfly.conf, the server also tries to load certificates from a file named butterfly_0.0.0.0.crt. Which interface I bind to,...

stat_monthly_user on cloud key v2 fails with self-signed certificate curl error

2

I'm using a UnifiOS device, the cloud key v2 pro, and talking to it using localhost (installing/running PHP on the device itself.) Many test scripts work fine. I'm trying the...

Using ZoneContextManager and FetchInstrumentation in browser causes lots of leaked Promises

13

### What version of OpenTelemetry are you using? "@opentelemetry/api": "^1.1.0", "@opentelemetry/context-zone": "^1.3.0", "@opentelemetry/core": "^1.3.0", "@opentelemetry/exporter-trace-otlp-http": "^0.29.1", "@opentelemetry/instrumentation-fetch": "^0.29.1", "@opentelemetry/sdk-trace-base": "^1.3.0", "@opentelemetry/semantic-conventions": "^1.3.0", ### What version of Node are you using?...

Make an unedited sample not create a gist

1

Pre-populating the box means that just pushing the button creates a bad gist. Instead, use placeholder to show the instructions.

Fix the build for M1 MacBooks

4

It turns out, MACOS is a perfectly fine OS to support aarch64 (through its name arm64.) The only thing missing was the appropriate check in impl_aarch64, plus renaming the file...

list_cpu_feature doesn't build on Apple M1

2

Note that Apple no longer sells new x64 MacBooks. It's literally impossible to get a new x64 MacBook for an engineer joining a team. Using CMake 3.22 When doing a...

Please build oboe as a single-header library

7

Issue #1072 was closed, but didn't actually address the feature request, seemingly from a mis-understanding of what a "single-header library" is. A "single-header library" is a library that is trivial...

Please document how to fine tune mpt-7b with longer contexts

3

## 🚀 Feature Request I believe that, to infer using a longer context, I can set max_seq_len to something longer when starting the huggingface based inference driver. However, I don't...

mpt-7b generates no output when finetuned with 2.0.1 based llm-foundry, works with 1.13.1

3

## Environment I'm using the Docker images for llm-foundry, training on 8xA100. mosaicml/llm-foundry:1.13.1_cu117-latest mosaicml/llm-foundry:2.0.1_cu118-latest Collecting system information... --------------------------------- System Environment Report Created: 2023-06-24 17:52:44 UTC --------------------------------- PyTorch information ------------------- PyTorch...

Composer container image doesn't show exceptions in logs when Trainer crashes

3

** Environment ** Collecting system information... --------------------------------- System Environment Report Created: 2023-05-24 19:24:21 UTC --------------------------------- PyTorch information ------------------- PyTorch version: 1.13.1+cu117 Is debug build: False CUDA used to build PyTorch:...