smollm2-135M-SFT-Only

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

smollm2-135M-8k-lc100k-mix1-ep2

This model is a fine-tuned version of HuggingFaceTB/SmoLLM2-135M on Smol-SmolTalk (the HuggingFaceTB/magpie-ultra-v1.0-top-300K-short-H4, the HuggingFaceTB/OpenHermes-2.5-H4-200k, the HuggingFaceTB/ifeval-like-data-36k-H4, the HuggingFaceTB/everyday-conversations-llama3.1-2k, the HuggingFaceTB/self-oss-instruct-sc2-H4, the HuggingFaceTB/summarization-data-10k-H4, the HuggingFaceTB/smollm-v2-summarization, the HuggingFaceTB/smollm-v2-rewriting-50k-H4, the HuggingFaceTB/explore-instruct-rewrite-H4 and the HuggingFaceTB/LongAlign-16k-ctx-english-H4 datasets). It achieves the following results on the evaluation set:

• Loss: 1.8390

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Framework versions

• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1