CoolFace
Modelpublic

RichardErkhov/amang1802_-_Llama3.2-1B-summary-length-exp2-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes705downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama3.2-1B-summary-length-exp2 - GGUF

  • —Model creator: https://huggingface.co/amang1802/
  • —Original model: https://huggingface.co/amang1802/Llama3.2-1B-summary-length-exp2/
NameQuant methodSize
Llama3.2-1B-summary-length-exp2.Q2_K.ggufQ2_K0.54GB
Llama3.2-1B-summary-length-exp2.IQ3_XS.ggufIQ3_XS0.58GB
Llama3.2-1B-summary-length-exp2.IQ3_S.ggufIQ3_S0.6GB
Llama3.2-1B-summary-length-exp2.Q3_K_S.ggufQ3KS0.6GB
Llama3.2-1B-summary-length-exp2.IQ3_M.ggufIQ3_M0.61GB
Llama3.2-1B-summary-length-exp2.Q3_K.ggufQ3_K0.64GB
Llama3.2-1B-summary-length-exp2.Q3_K_M.ggufQ3KM0.64GB
Llama3.2-1B-summary-length-exp2.Q3_K_L.ggufQ3KL0.68GB
Llama3.2-1B-summary-length-exp2.IQ4_XS.ggufIQ4_XS0.7GB
Llama3.2-1B-summary-length-exp2.Q4_0.ggufQ4_00.72GB
Llama3.2-1B-summary-length-exp2.IQ4_NL.ggufIQ4_NL0.72GB
Llama3.2-1B-summary-length-exp2.Q4_K_S.ggufQ4KS0.72GB
Llama3.2-1B-summary-length-exp2.Q4_K.ggufQ4_K0.75GB
Llama3.2-1B-summary-length-exp2.Q4_K_M.ggufQ4KM0.75GB
Llama3.2-1B-summary-length-exp2.Q4_1.ggufQ4_10.77GB
Llama3.2-1B-summary-length-exp2.Q5_0.ggufQ5_00.83GB
Llama3.2-1B-summary-length-exp2.Q5_K_S.ggufQ5KS0.83GB
Llama3.2-1B-summary-length-exp2.Q5_K.ggufQ5_K0.85GB
Llama3.2-1B-summary-length-exp2.Q5_K_M.ggufQ5KM0.85GB
Llama3.2-1B-summary-length-exp2.Q5_1.ggufQ5_10.89GB
Llama3.2-1B-summary-length-exp2.Q6_K.ggufQ6_K0.95GB
Llama3.2-1B-summary-length-exp2.Q8_0.ggufQ8_01.23GB

Original model description: --- library_name: transformers tags: [] ---

Model Card for Model ID

  • —Summary Length PPO experiment #2
  • —No KL divergence in loss

Model Details

  • —Dataset size: 1024
  • —Epochs: 2
  • —Batch Size: 4 * 8 (using Grad Accu)

Optimizer args: Torch AdamW default, except

  • —LR = 0.0001