Spaces:
Sleeping
Sleeping
Commit History
-
adds a lot of improvements to launch.sh and demo spaces and secures read token at demo launch
36b8703 -
improves spaces deployment , configuration for custom settings , adds interface for spaces deployment
5f8b28d
-
propagates changes
468cb10 -
adds improved dataset utils with new interface
b172ca3 -
solves import error
c238a70 -
adds imports and forces monitoring
190d843 -
adds scheduler stuff and hopes for the best with track tonic
d455d12 -
adds local and remote training monitors to config
71db310 -
improve dataset utils non destructive writes
924581c -
adds improved launch for reasoning gpt-oss configs and new config for medical reasoning
0ded6bb -
adds optimizations for faster training
3331c7f -
adds a100 memory optimized
fa9560d -
adds template files , adds non destructive dataset updates
d47568c -
adds flash attention 3 kernel
cb276d8
-
adapt lr_scheduler according to trl version
976e218 -
sets min_lr
7f45871 -
adds better launch.sh and eval / test splits auto
0fa6045 -
adds repoid only based on repo name, adds version-robust sfttrainer
665844a -
coerce akk numeric config values to safe values
c346dad -
adds defensive programming (boo) and adaptations based on transformer versions
97dacc7 -
improves launcher with model family and defaults based on options, updates trl trainer , removes trl config paths by switching to trainingarguments class , tokenizer parameters updated to sfttrainer , resolves evaluation_strategy error
598357a -
hide all tokens in logs, never persist to disk, remove max_seq_length from config, add to trainer
eb9e91f -
adds pythonpath based on chatgpt5 suggestion
c3ab72f
-
adds harmony format , configurable gpt-oss parameters, launch.sh logic , improved templates for legml gpt-oss training, dynamic results directory and improve model pushing
59e57ff -
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
75bcdb3
-
adds quantization configuration correctly
c7cffbb -
adds quantization configuration correctly
fa7de39 -
adds memory optimized configuration
7181190 -
fixes custom trackio implementation
dfcb060 -
adds gpt-oss support
fcf2981
-
adds readme, removes quantization, adds readtoken logic, updates trackio , spaces
3c37508
-
cleanup a bit the files
Unverified signaturead3b15d -
workaround for quantization and push
TonicUnverified signature2432208 -
workaround for quantization and push
TonicUnverified signatureb79fab9 -
solves model card formatting bug
TonicUnverified signature41e9e02 -
solves model card formatting bug
TonicUnverified signature2f866e6 -
Fix model recovery and deployment scripts - add safetensors support and Windows compatibility
TonicUnverified signatured0d19b2
-
resolves the urls correctly
TonicUnverified signaturec3f29a5 -
use gradio for better connection with the spaces
TonicUnverified signaturef251d3d -
adds monkey patch for trackio monitoring in torch and readme creator improvements
TonicUnverified signature39db0ca
-
fixes linter errors in setup hf dataset
TonicUnverified signature2df26a0 -
fixes monitoring
TonicUnverified signaturec61ed6b -
adds automation for hf cli using token
TonicUnverified signature5d7656c -
adds hf cli fixes
TonicUnverified signaturefd0524b -
adds new hf cli
TonicUnverified signatured291e63 -
solves hf cli error
TonicUnverified signaturec2321bb -
improves model card
TonicUnverified signature3eb616f -
adds sft , quantization, better readmes
TonicUnverified signature40fd629
-
attempts to resolve deployment issues
TonicUnverified signaturec417358 -
adds better huggingface deploy
TonicUnverified signature2da5c04