| { |
| "method": "gemma_to_olmo_affine_ridge", |
| "teacher": "/workspace/huggingface/hub/models--google--gemma-3-12b-pt/snapshots/295efb63d01a7017928f273a94ebb86105c9526f", |
| "student": "allenai/OLMo-2-0425-1B", |
| "expanded_tokenizer": "/workspace/ugrip-kd/research/results/expansion/tokenizer", |
| "vocab_map": "/workspace/ugrip-kd/research/results/expansion/tokenizer/vocab_map.json", |
| "anchor_pool": "english", |
| "anchor_count": 60900, |
| "validation_count": 6090, |
| "validation_metrics": { |
| "input": { |
| "rmse": 0.21349406242370605, |
| "mean_cosine": 0.5147131085395813, |
| "median_cosine": 0.5201483964920044 |
| }, |
| "output": { |
| "rmse": 0.023339398205280304, |
| "mean_cosine": 0.8253485560417175, |
| "median_cosine": 0.8286930322647095 |
| } |
| }, |
| "ridge": 0.001, |
| "base_vocab_size": 100278, |
| "student_embedding_rows_before_resize": 100352, |
| "padded_rows_before_resize": 74, |
| "new_token_count": 7212, |
| "new_token_teacher_coverage": 7212, |
| "seed": 42 |
| } |
|
|
