Stablelm 2 12B Chat is an open-source language model by stabilityai. Features: 12b LLM, VRAM: 24.4GB, Context: 4K, License: other, Instruction-Based, LLM Explorer Score: 0.18, Arc: 64.9, HellaSwag: 86, MMLU: 61.1, GSM8K: 57.9.
Stablelm 2 12B Chat Parameters and Internals
| Model Type | |
| Use Cases |
| Areas: | |
| Limitations: | | requires guardrails around inputs and outputs to prevent harmful responses, not suitable for applications that may cause harm |
|
|
| Supported Languages | |
| Training Details |
| Data Sources: | | HuggingFaceH4/ultrachat_200k, allenai/ultrafeedback_binarized_cleaned, meta-math/MetaMathQA, WizardLM/WizardLM_evol_instruct_V2_196k, openchat/openchat_sharegpt4_dataset, LDJnr/Capybara, Intel/orca_dpo_pairs, hkust-nlp/deita-10k-v0, Anthropic/hh-rlhf, glaiveai/glaive-function-calling-v2, Internal Safety Dataset |
|
| Methodology: | | Direct Preference Optimization (DPO) |
|
| Model Architecture: | | transformer decoder architecture |
|
|