212 1 40

Ronan McGovern

RonanMcGovern

https://ronanmcgovern.com

AI & ML interests

Open source LLMs. Fine-tuning. Summarisation. Patents.

Recent Activity

updated a model 1 day ago

Trelis/transcribe-en_us-spelling-v1-turbo-ctranslate2

updated a model 1 day ago

Trelis/transcribe-en_us-spelling-v1-tiny-ctranslate2

updated a model 1 day ago

Trelis/transcribe-en_us-spelling-v1-turbo

View all activity

Organizations

New activity in Trelis/TRM-ARC-AGI-II about 1 month ago

Thanks !

#1 opened about 1 month ago by

cdtmc

New activity in kyutai/stt-1b-en_fr 3 months ago

What is the tokenization and alignment approach? i.e. collation

#9 opened 3 months ago by

RonanMcGovern

New activity in openai/gpt-oss-20b 3 months ago

Does transformers utilize PyTorch SDPA's flash_attention for openai/gpt-oss-20b?

#89 opened 4 months ago by

NooBaymax

New activity in transformers-community/support 3 months ago

Clarification on Recent Changes to Loss and Gradient Accumulation

👍 2

#15 opened 5 months ago by

jiosephlee

New activity in mistralai/Voxtral-Mini-3B-2507 5 months ago

Fine tuning of Model

#22 opened 5 months ago by

amritansh

New activity in lerobot/smolvla_base 5 months ago

Data norm issue during evaluation/inference

#8 opened 5 months ago by

RonanMcGovern

New activity in amazon/chronos-t5-small 7 months ago

Amazon Chronos T5 Small Forecasting Google Colab Notebook

👍 1

#4 opened over 1 year ago by

JamesBentley

New activity in Trelis/Mixtral-8x7B-Instruct-v0.1-function-calling-v3 7 months ago

Ollama with GGUF

#6 opened 8 months ago by

smaram

New activity in Qwen/Qwen3-30B-A3B-FP8 7 months ago

SGLang very slow ~6 toks with 1 concurrency on H100SXM

#3 opened 7 months ago by

RonanMcGovern

New activity in leon-se/gemma-3-27b-it-FP8-Dynamic 8 months ago

Low inference throughput?

#2 opened 8 months ago by

RonanMcGovern

New activity in Qwen/Qwen2.5-Omni-7B 8 months ago

Executing Qwen2.5-Omni-7B on SGLang: AttributeError: 'Qwen2_5OmniConfig' object has no attribute 'hidden_size'

#21 opened 9 months ago by

didduran

New activity in google/gemma-3-4b-it 9 months ago

Model Produces no Outputs

#16 opened 9 months ago by

RonanMcGovern

New activity in Qwen/Qwen2-VL-7B-Instruct 10 months ago

[BUG] {'use_reentrant': True} results in "Gradients will be None"

#74 opened 11 months ago by

RonanMcGovern

New activity in onnx-community/Qwen2.5-Coder-1.5B-Instruct 10 months ago

Model does not follow or acknowledge system prompts?

#3 opened 10 months ago by

RonanMcGovern

New activity in onnx-community/Phi-3.5-mini-instruct-onnx-web 10 months ago

Transformers 3.0 can't find files

#5 opened about 1 year ago by

r0-0rd

New activity in vikhyatk/moondream2 11 months ago

Unable to load finetuned model after saving

#57 opened 11 months ago by

Charlington

New activity in jxm/cde-small-v2 11 months ago

Set base_model & tags metadata

#1 opened 11 months ago by

tomaarsen

New activity in vikhyatk/moondream2 11 months ago

Finetuning for pointing, object detection task

#48 opened 11 months ago by

yaneivan

New activity in RedHatAI/Sparse-Llama-3.1-8B-2of4 12 months ago

Can I apply a LoRA?

#1 opened 12 months ago by

RonanMcGovern

New activity in Qwen/Qwen2-Audio-7B-Instruct about 1 year ago

Is vLLM support planned?

#10 opened about 1 year ago by

RonanMcGovern

Ronan McGovern

AI & ML interests

Recent Activity

Organizations

RonanMcGovern's activity

Thanks !

What is the tokenization and alignment approach? i.e. collation

Does transformers utilize PyTorch SDPA's flash_attention for openai/gpt-oss-20b?

Clarification on Recent Changes to Loss and Gradient Accumulation

Fine tuning of Model

Data norm issue during evaluation/inference

Amazon Chronos T5 Small Forecasting Google Colab Notebook

Ollama with GGUF

SGLang very slow ~6 toks with 1 concurrency on H100SXM

Low inference throughput?

Executing Qwen2.5-Omni-7B on SGLang: AttributeError: 'Qwen2_5OmniConfig' object has no attribute 'hidden_size'

Model Produces no Outputs

[BUG] {'use_reentrant': True} results in "Gradients will be None"

Model does not follow or acknowledge system prompts?

Transformers 3.0 can't find files

Unable to load finetuned model after saving

Set base_model & tags metadata

Finetuning for pointing, object detection task

Can I apply a LoRA?

Is vLLM support planned?