{ "fixture": "huggingface-mic/luke-base-mic", "key": "c9b636bf3b64505d3274eb4a79f595e0296b17ee3d63948b2a9c4856b110112b", "key_fields": { "architectures": [ "LukeForMaskedLM" ], "output_hidden_states": false, "return_dict": true, "chunk_size_feed_forward": 0, "is_encoder_decoder": false, "id2label": { "0": "LABEL_0", "1": "LABEL_1" }, "label2id": { "LABEL_0": 0, "LABEL_1": 1 }, "problem_type": null, "vocab_size": 50267, "entity_vocab_size": 500000, "hidden_size": 64, "entity_emb_size": 6, "num_hidden_layers": 2, "num_attention_heads": 4, "intermediate_size": 128, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "attention_probs_dropout_prob": 0.1, "max_position_embeddings": 514, "type_vocab_size": 1, "initializer_range": 0.02, "layer_norm_eps": 1e-05, "use_entity_aware_attention": true, "classifier_dropout": null, "pad_token_id": 1, "bos_token_id": 0, "eos_token_id": 2, "tie_word_embeddings": true, "bert_model_name": "roberta-base", "gradient_checkpointing": false, "model_type": "luke", "output_past": true, "position_embedding_type": "absolute", "output_attentions": false }, "covers": { "studio-ousia/luke-base": "7438924defd9f3c2018d63c16073bf4bcb6a70aa" }, "built_from": { "repo": "studio-ousia/luke-base", "revision": "7438924defd9f3c2018d63c16073bf4bcb6a70aa" }, "seed": 17, "dtype": "float32", "builder": "example/build_fixtures.py", "builder_revision": "e8034a51dc828ccfb8c6ad3d765dd2275181f0b8", "recipe": { "seed": 17, "model_class": "LukeForMaskedLM" }, "shrink": "rule", "model_class": "LukeForMaskedLM", "parameters": 10151757, "missing_parameters": [ "lm_head.decoder.weight" ], "serialization": { "file": "model.safetensors", "reverted_conversions": [], "source_file": "pytorch_model.bin", "migrated_from": "pytorch_model.bin" }, "tester_sizes": { "source": "https://raw.githubusercontent.com/huggingface/transformers/main/tests/models/luke/test_modeling_luke.py", "fields": { "entity_emb_size": 6 } }, "checkpoint_inspection": "parameter headers/indexes; bounded binary fallback", "numeric_health_scope": "CPU float32/bfloat16/float16; autocast for low-precision losses", "captured": { "call": "training step", "inputs": [ "attention_mask", "entity_attention_mask", "entity_ids", "entity_position_ids", "entity_token_type_ids", "input_ids", "labels" ], "class": "LukeForMaskedLM" }, "weights": "random", "gate": { "random": { "pass": true, "accepted_insensitive": {}, "threshold": 0.001, "deltas": { "input_ids:roll_batch": 0.76102614402771, "input_ids:reverse": 0.924561619758606, "attention_mask:all_ones": 0.003961771726608276 }, "weakest": "attention_mask:all_ones", "no_perturbation_defined": [ "entity_ids", "entity_attention_mask", "entity_token_type_ids", "entity_position_ids" ], "not_a_forward_parameter": [] } }, "numeric_health": { "float32": { "outputs_finite": true, "loss_finite": true, "gradients_finite": true, "finite": true }, "bfloat16": { "outputs_finite": true, "loss_finite": true, "gradients_finite": true, "finite": true }, "float16": { "outputs_finite": true, "loss_finite": true, "unscaled_gradients_finite": true, "finite": true } }, "copied_unchanged": [ "added_tokens.json", "entity_vocab.json", "merges.txt", "special_tokens_map.json", "tokenizer_config.json", "vocab.json" ], "environment": { "python": "3.12.12", "machine": "arm64", "torch": "2.14.1", "transformers": "5.22.0.dev0", "tokenizers": "0.23.3" }, "sha256": { "added_tokens.json": "9c8d4b2cd7d3fc675213fd88eb40a6308e180d1d95dee49d51bc716daa7871cf", "config.json": "a125bdec094fae0be62ac63e00eb3fd407978370e076e413217ec18de7e696a1", "entity_vocab.json": "45e0508babfee0667db9a8dfa2519a6e470c769c3489b4c87c45b26f59e30947", "merges.txt": "1ce1664773c50f3e0cc8842619a93edc4624525b728b188a9e0be33b7726adc5", "model.safetensors": "2138364362527516537719e0ecb87fc0beef3fc8cb653d5161fe2dc718bd2c25", "special_tokens_map.json": "612f89fad4371cf030156e51c80c5a6e570299052a6c7dc2c958f34c4c34e2b8", "tokenizer_config.json": "f51c09c337e73afe1c03325bd842b96ad9dbe1e27a0eacea590b4a240f2b8eba", "vocab.json": "06b4d46c8e752d410213d9548eb27a54db70fda0319b6271fb8d59dead5e1cab" } }