Autotrain compatible Base model:finetune:floflob/sm... Base model:floflob/smollm2-135... Endpoints compatible Generated from trainer Llama Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/FlofloB/smollm2-135M_pretrained_1400k_fineweb
Smollm2 135M Pretrained 1400k Fineweb Benchmarks
MMLU Pro: 0.88
GPQA: 2.13
MUSR: 6.02
BBH: 2.16
IFEval: 17.64 vs 88 (so35)-80%
MATH Lvl 5: 1.13
LLME Score: 0.23615
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Smollm2 135M Pretrained 1400K Fineweb Parameters and Internals
| LLM Name | Smollm2 135M Pretrained 1400k Fineweb |
| Repository 🤗 | https://huggingface.co/FlofloB/smollm2-135M_pretrained_1400k_fineweb |
| Base Model(s) | |
| Model Size | 135m |
| Required VRAM | 0.5 GB |
| Updated | 2025-05-21 |
| Maintainer | FlofloB |
| Model Type | llama |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 8192 |
| Model Max Length | 8192 |
| Transformers Version | 4.44.2 |
| Tokenizer Class | GPT2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 49152 |
| Torch Data Type | float32 |
Best Alternatives to Smollm2 135M Pretrained 1400K Fineweb
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| SmolLM2 135M Instruct | 8K / 0.3 GB | 332358 | 194 |
| SmolLM2 135M | 8K / 0.3 GB | 556758 | 94 |
| SmolLM2 135M Eagle | 8K / 0.3 GB | 64 | 3 |
| SmolLM2 FT MyDataset | 8K / 0.5 GB | 65 | 0 |
| SmolLM2 135M Instruct Ita | 8K / 0.1 GB | 31 | 0 |
| Sft Output | 8K / 0.5 GB | 9 | 0 |
| SmolLM2 135M Instruct | 8K / 0.3 GB | 1769 | 3 |
| SmolLM2 FT Eng | 8K / 0.5 GB | 7 | 0 |
| ... Fineweb Uncovai Human Removed | 8K / 0.5 GB | 7 | 0 |
| SmolLM2 135M Grpo Gsm8k | 8K / 0.5 GB | 42 | 7 |
Note: green Score (e.g. "73.2") means that the model is better than FlofloB/smollm2-135M_pretrained_1400k_fineweb .
Rank the Smollm2 135M Pretrained 1400K Fineweb Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 47473 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227