consolidate: drop the duplicated root copy

README.md

@@ -1,43 +0,0 @@
----
-library_name: transformers
-tags: [model-merging, mergeability, training-free, quotient-merge-distance]
----
-
-# goldfish-en-spa_latn__average__naive
-
-Training-free merged checkpoint from the **Mergeability** sweep
-(`benchmark/emit_lm.py --real`), produced by weight-space merging of two independently
-trained parents. No gradient steps were taken.
-
-| field | value |
-|---|---|
-| pair_id | `goldfish-en-spa_latn` |
-| parent_a | `goldfish-models/eng_latn_1000mb` |
-| parent_b | `goldfish-models/spa_latn_1000mb` |
-| ceiling | `catherinearnett/B-GPT_en_es_simultaneous` |
-| operator | `average` |
-| alignment | `naive` |
-| align_method | `permutation` |
-| regime | `different_corpora` |
-| eval_langs | `eng_latn+spa_latn` |
-| nll_merge | `8.6857` |
-| nll_floor | `4.5501` |
-| param_coverage | `1.0` |
-| MS | `-6.081` |
-
-## How it was made
-
-Parents were loaded, activations extracted on a shared calibration corpus, and the merge applied
-either **naive** (parents combined in their own coordinates) or **aligned** (parent B carried into
-parent A's residual-stream basis via `common.alignment.residual_basis_map` before merging —
-permutation for same-width pairs, orthogonal/rectangular for cross-width).
-
-`MS` is the recovery score from `common.eval.mergeability_score` (merged vs. floor vs. ceiling), the
-same normalisation used by Zhou et al., so it is comparable across rows of the sweep.
-
-## Caveats
-
-Sub-1B merges are noisy; an aligned signal where the naive one is noise is the finding, not a bug.
-Rows without a joint ceiling are floor-relative and must not be read as absolute recovery.
-
-Generated automatically — see the [mergeability repo](https://github.com/suchirsalhan/mergeability).

config.json

@@ -1,36 +0,0 @@
-{
-  "activation_function": "gelu",
-  "add_cross_attention": false,
-  "architectures": [
-    "GPT2LMHeadModel"
-  ],
-  "attn_pdrop": 0.1,
-  "bos_token_id": 50000,
-  "dtype": "float32",
-  "embd_pdrop": 0.1,
-  "eos_token_id": 50001,
-  "initializer_range": 0.02,
-  "layer_norm_epsilon": 1e-05,
-  "model_type": "gpt2",
-  "n_ctx": 512,
-  "n_embd": 768,
-  "n_head": 12,
-  "n_inner": 3072,
-  "n_layer": 12,
-  "n_positions": 512,
-  "pad_token_id": 50002,
-  "prefix": "[CLS]",
-  "reorder_and_upcast_attn": false,
-  "resid_pdrop": 0.1,
-  "scale_attn_by_inverse_layer_idx": false,
-  "scale_attn_weights": true,
-  "summary_activation": null,
-  "summary_first_dropout": 0.1,
-  "summary_proj_to_labels": true,
-  "summary_type": "cls_index",
-  "summary_use_proj": true,
-  "tie_word_embeddings": true,
-  "transformers_version": "5.14.1",
-  "use_cache": true,
-  "vocab_size": 51200
-}

generation_config.json

@@ -1,8 +0,0 @@
-{
-  "_from_model_config": true,
-  "bos_token_id": 50000,
-  "eos_token_id": 50001,
-  "pad_token_id": 50002,
-  "transformers_version": "5.14.1",
-  "use_cache": true
-}

model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d9d491eccd8fb47af1e57fdc593a225945c411e84ba7bc502d6e8f24bab08dc7
-size 499098240

tokenizer_config.json

@@ -1,12 +0,0 @@
-{
-  "backend": "tokenizers",
-  "bos_token": "[CLS]",
-  "cls_token": "[CLS]",
-  "eos_token": "[SEP]",
-  "mask_token": "[MASK]",
-  "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "<pad>",
-  "sep_token": "[SEP]",
-  "tokenizer_class": "TokenizersBackend",
-  "unk_token": "<unk>"
-}