ashrafulparan/llama-3-finetuned-for-subjectivity-english-lora

ashrafulparan/llama-3-finetuned-for-subjectivity-english-lora

<!-- Provide a quick summary of what the model is/does. --> It is a finetuned version of llama-3 for subjectivity dataset. It will output 1 for Subjectivity and 0 for Objectivity.

Model Details

Model Hyperparameters

args = TrainingArguments( perdevicetrainbatchsize = 2, gradientaccumulationsteps = 4, warmupsteps = 5, numtrainepochs = 12, learningrate = 5e-5, fp16 = not torch.cuda.isbf16supported(), bf16 = torch.cuda.isbf16supported(), loggingsteps = 10, optim = "adamw8bit", weightdecay = 0.001, lrschedulertype = "linear", seed = 3407, outputdir = "outputs", report_to = "none",)

<!-- Provide a longer summary of what this model is. -->

Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

BibTeX:

[More Information Needed]

Framework versions

• —PEFT 0.11.1