DEV Community: Trevor Davis

What actors appeared in both Jupiter Ascending and Game of Thrones?

Trevor Davis — Fri, 28 Feb 2025 14:12:13 +0000

There I was lying in bed last night trying to form a connecting between the Seals and Signets Minister in Jupiter Ascending and Grand Maester Pycell in Game of Thrones. My tired brain thought they might be the same actor who portrayed both characters. They are not. Terry Gilliam is the former, Julian Glover the latter.

I decided to look this information up this morning. I could have gone to IMDB or Wikipedia to read about the film and show and their respective casts. This is 2025, so I gave AI a shot at it first.

First up, chat.com from OpenAI.

Not a great start. Chat decides Edward Hogg was in the movie and the show. Edward is indeed credited in Jupiter Ascending. He was not credited in Game of Thrones. Moreover, the specificity of the answer leads me to believe this is an indication that a hallucination took place. The final sentence also contradicts itself sating there are no overlaps, but Edward Hogg is a notable connection.

Next up, perplexity.com. I had high hopes for the answer from perplexity.

I want to love these new search engine services like perplexity. This kind of a response gives me pause. The first line, "There are no actors who appeared in both Jupiter Ascending and Game of Thrones" is already incorrect. Worse still, In the second paragraph, perplexity gets the correct answer, Sean Bean, though does not in fact make the connection. The paragraph ends with "However, he is not a common actor between the two series." The prose ends with "Therefore, there are no actors who have appeared in both Jupiter Ascending and Game of Thrones." Quickly scanning these results as an information worker might do during research would lead you to the wrong conclusion. Disappointing.

Finally, I tried some local AI. I used Ollama v0.5.7 and llama3.3-latest from the model repository.

Llama is the first model to get the correct answer of Sean Bean, including the characters he played in the movie and the show. It should have stopped here.

Llama also hallucinated. It first gave Gugu Mbatha-Raw as a second answer. Gugu Mbatha-Raw is in Jupiter Ascending, but not Game of Thrones. Interestingly, throw the course of answering, Llama realizes this, and hallucinates another connection, David Ajala, who is also in Jupiter Ascending but not Game of Thrones.

Generated text from large language models have their uses. For reference information, I am going to stick with curated sources for reliable information for the time being.

Cover art generated using Adobe® Firefly

Ollama Environment Variables

Trevor Davis — Wed, 29 Jan 2025 13:18:28 +0000

If you're writing an integration with Ollama or want to influence its function on your device, there are several environment variables you can set to configure how Ollama operates.

Source (2025-01-29): https://github.com/ollama/ollama/blob/main/envconfig/config.go

OLLAMA_HOST- defaults to http://127.0.0.1:11434. Can be http/https. Host returns the scheme and host. Host can be configured via the OLLAMA_HOST environment variable.
OLLAMA_ORIGINS- comma separated values. Always appends http(s)://localhost, 127.0.0.1, 0.0.0.0, app://, file://, tauri://, vscode-webview://. Origins returns a list of allowed origins. Origins can be configured via the OLLAMA_ORIGINS environment variable.
OLLAMA_MODELS- Models returns the path to the models directory. Models directory can be configured via the OLLAMA_MODELS environment variable. Default is $HOME/.ollama/models
OLLAMA_KEEP_ALIVE- KeepAlive returns the duration that models stay loaded in memory. KeepAlive can be configured via the OLLAMA_KEEP_ALIVE environment variable. Negative values are treated as infinite. Zero is treated as no keep alive. Default is 5 minutes.
OLLAMA_LOAD_TIMEOUT- LoadTimeout returns the duration for stall detection during model loads. LoadTimeout can be configured via the OLLAMA_LOAD_TIMEOUT environment variable. Zero or Negative values are treated as infinite. Default is 5 minutes.

Boolean flags. It accepts 1, t, T, TRUE, true, True, 0, f, F, FALSE, false, False. Any other value returns an error. (https://pkg.go.dev/strconv#ParseBool)

OLLAMA_DEBUG - Debug enabled additional debug information.
OLLAMA_FLASH_ATTENTION - FlashAttention enables the experimental flash attention feature.
OLLAMA_KV_CACHE_TYPE - KvCacheType is the quantization type for the K/V cache.
OLLAMA_NOHISTORY - NoHistory disables readline history.
OLLAMA_NOPRUNE - NoPrune disables pruning of model blobs on startup.
OLLAMA_SCHED_SPREAD - SchedSpread allows scheduling models across all GPUs.
OLLAMA_INTEL_GPU - IntelGPU enables experimental Intel GPU detection.
OLLAMA_MULTIUSER_CACHE - MultiUserCache optimizes prompt caching for multi-user scenarios

Library Settings

OLLAMA_LLM_LIBRARY - Set LLM library to bypass autodetection
CUDA_VISIBLE_DEVICES - Set which NVIDIA devices are visible
HIP_VISIBLE_DEVICES - Set which AMD devices are visible by numeric ID
ROCR_VISIBLE_DEVICES - Set which AMD devices are visible by UUID or numeric ID
GPU_DEVICE_ORDINAL - Set which AMD devices are visible by numeric ID
HSA_OVERRIDE_GFX_VERSION - Override the gfx used for all detected AMD GPUs

Performance Settings

OLLAMA_NUM_PARALLEL - Maximum number of parallel requests. Defaults to unlimited
OLLAMA_MAX_LOADED_MODELS - Maximum number of loaded models per GPU. Defaults to unlimited.
OLLAMA_MAX_QUEUE - Maximum number of queued requests. Defaults to 512.
OLLAMA_MAX_VRAM - Maximum amount of VRAM that can be consumed per GPU. Default is unlimited
OLLAMA_GPU_OVERHEAD - Set aside VRAM per GPU. Default is 0

Proxy Settings

HTTP_PROXY - HTTP proxy
HTTPS_PROXY - HTTPS proxy
NO_PROXY - No proxy

netsh int ip reset - for when you don't need to know what the problem was

Trevor Davis — Sun, 02 Feb 2020 22:55:20 +0000

There is something to be said for finding the problem and fixing it. And there is also something to be said for hitting it with a hammer until it works/fits.

I use this command in the latter category.

netsh is an amazing command that is a complete toolbox for the Windows networking stack covering most areas of functionality in numbing nuance. It is for this reason that when things go wrong, this is the command that will fix it or tell you what is going on.

That's all well and good, except you don't always have time and patience for finding the problem, and often, a quick fix is just what you need.

netsh int ip reset is a mostly safe reset of the whole network stack. I use it to fix connection problems with WiFi, DHCP lease problems, and VPN connections that refuse to connect. After a quick reboot, the issues are mostly fixed for me.

This magical command isn't going to fix any real problems like an unplugged cable for an internet issue; it's good for those times when things should be working and you can't see why. Use it like a hammer.

Hello World

Trevor Davis — Thu, 23 Jan 2020 22:13:08 +0000

This is my first post.