HuggingFaceTB/SmolLM-135M-Instruct
Full fine-tune vs HuggingFaceTB/SmolLM-135M. Fine-tune: 100% of the weights changed, across 30 of 30 layers.
Tensors changed: Embeddings 1/1, Norms 61/61, MLP 90/90, Attention 120/120
15 derivatives checked
- HuggingFaceTB/smollm-135M-instruct-v0.2-Q8_0-GGUF: Quantization
- mradermacher/SmolLM-135M-Instruct-GGUF: Full fine-tune
- Felladrin/gguf-SmolLM-135M-Instruct: Full fine-tune
- second-state/SmolLM-135M-Instruct-GGUF: Full fine-tune
- MaziyarPanahi/SmolLM-135M-Instruct-GGUF: Full fine-tune
- mradermacher/SmolLM-135M-Instruct-i1-GGUF: Quantization
- gaianet/SmolLM-135M-Instruct-GGUF: Full fine-tune
- arzaan789/smollm-135m-uncensored: Full fine-tune
- unsloth/SmolLM-135M-Instruct: Identical copy
- RedHatAI/SmolLM-135M-Instruct-quantized.w8a16: Quantization
- unsloth/SmolLM-135M-Instruct-bnb-4bit: Quantization
- jtatman/llama_smollm_135m_skunkworks_reasoning_alpaca_gguf: Full fine-tune
- Liquid1/MSU: LoRA-style
- unclemusclez/SmolLM-135M-Instruct-DEVINator-v0.1: Same weights, converted
- Trelis/SmolLM-135M-Instruct-layer-pruned-90M-raw: Partial fine-tune (fewer layers)
10 other repos hold byte-for-byte identical weight files.