hide all tokens in logs, never persist to disk, remove max_seq_length from config, add to trainer

adds pythonpath based on chatgpt5 suggestion

adds harmony format , configurable gpt-oss parameters, launch.sh logic , improved templates for legml gpt-oss training, dynamic results directory and improve model pushing

adds single token logic read/write , adds gpt-oss demo space , adds spaces refactor , adds new version of track tonic , adds logic in launch.sh

adds quantization configuration correctly

adds quantization configuration correctly

adds memory optimized configuration

fixes custom trackio implementation

adds gpt-oss support

adds readme, removes quantization, adds readtoken logic, updates trackio , spaces

cleanup a bit the files

workaround for quantization and push

workaround for quantization and push

solves model card formatting bug

solves model card formatting bug

Fix model recovery and deployment scripts - add safetensors support and Windows compatibility

resolves the urls correctly

use gradio for better connection with the spaces

adds monkey patch for trackio monitoring in torch and readme creator improvements

fixes linter errors in setup hf dataset

fixes monitoring

adds automation for hf cli using token

adds hf cli fixes

adds new hf cli

solves hf cli error

improves model card

adds sft , quantization, better readmes

attempts to resolve deployment issues

adds better huggingface deploy

attempts spaces fix

adds correct huggingface spaces api deployment

Initial Space setup

fixes callback , deploy , and trainer bug

Initial Space setup

Initial Space setup

Merge branch 'main' of https://github.com/Josephrp/smollm3_finetune

fix config bug

Initial Space setup

Merge branch 'main' of https://github.com/Josephrp/smollm3_finetune

adds better huggingface deploy

Initial Space setup

fix launch script deploy and refactors

fix launch script

adds formatting fix