chore: update README and REPORT with performance insights and dataset changes

chore: update dependencies and configuration for improved training

chore: updated eval charts

chore: moved tokenlearn as in internal package

chore: moved model2vec as in internal package

feat(distiller): add option to skip post-training re-regularization

feat(analyze): enhance model type identification in analysis

refactor(report): reflect distillation experiments data in report

chore(analysis): update image files

fix(pretrain): clamp input IDs to prevent out of bounds error

chore(distiller): remove unused GPU_NAME variable

feat(config): allow multiple GPU types for training and simplify GPU handling

refactor(patch-utils): improve patch application check for tokenlearn

refactor(distiller): improve beam distillation and tokenlearn integration

docs(readme): refine dataset identifiers in README

docs: rename model to codemalt and update evaluation instructions

refactor(beam-utils): use direct file operations for beam volumes

feat(distiller): add checkpointing and refactor analyze.py

chore(model): update model weights

chore: add initial CodeMap configuration file

chore: remove MTEB results analysis script

chore: add ignore patterns for temp and env files

chore: configure git lfs for mteb and evaluation files

build: add taskfile for common commands

chore(analysis-charts): add batch size and memory scaling charts

chore(model): remove metadata file for distilled model

chore(analysis-charts): add peer comparison chart

chore(analysis-charts): add benchmark performance chart

chore(assets): add language heatmap image

fix(main): reduce maximum queries per language to 100

chore(deps): update dependencies in uv.lock

chore(analysis-charts): add new radar charts for code model embeddings

feat(distiller): configure beam functions with resource settings

feat: introduce distiller package and update README

feat(config): add new model type for vector embeddings

chore(analysis-charts): add new image assets

chore: remove unused scripts and update dependencies

feat: overhaul distiller package with unified CLI, enhanced evaluation, and modular structure

feat: created a cli to manage the complete generation process

feat: 4 stage training, refinement failed for first 3

feat: coding stage training completed

feat: potion-based trained model

chore: add env vars

feat: added MTEB evaluation scripts

fix: migrate binary files to LFS tracking

chore: set lfs tracking

docs: Enhance README.md with detailed project information for gte-Qwen2-7B-instruct-M2V-Distilled, including model optimization benefits, installation instructions, usage examples, and performance results.

initial commit

initial commit