The temporary token-level regression head used during training is not included. Load the checkpoint with AutoModel.from_pretrained for downstream encoder analysis or continued fine-tuning.

python

from transformers import AutoModel, AutoTokenizer

model_id = "calogero-jerik-scozzaro/BERT_MULTIPLEYE_no_HumanRights_seed_46"
tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True)
model = AutoModel.from_pretrained(model_id)

Training metadata

Field	Value
`all_train_texts`	Arg_PISACowsMilk, Arg_PISARapaNui, LearningMobility, Lit_Alchemist, Lit_BrokenApril, Lit_MagicMountain, Lit_Solaris, PopSci_Caveman, PopSci_MultiplEYE
`batch_size`	8
`epochs`	100
`learning_rate`	2e-05
`max_length`	256
`measures`	FFD, FPRT, TFT, RRT, skipped, FPF, RR
`num_train_sentences`	90
`seed`	46
`source_model`	dbmdz/bert-base-italian-uncased
`stage`	1
`stage_train_texts`	Arg_PISACowsMilk, Arg_PISARapaNui, LearningMobility, Lit_Alchemist, Lit_BrokenApril, Lit_MagicMountain, Lit_Solaris, PopSci_Caveman, PopSci_MultiplEYE
`test_texts`	HumanRights
`variant`	BERT_MULTIPLEYE_no_HumanRights

The uploaded files include et_label_scaler.json, which records the min-max scaling statistics used for the eye-tracking labels.