300 GB
10,168 files
Updated 18 days ago
Name
Size
pretrain-hc-1600M-L28-lr1e3-b200-8gpu-segmented
pretrain-hc-1600M-L28-lr5e4-b200-8gpu-segmented-wandb
pretrain-hc-200M-L12-lr2e3-ladder-20260914
pretrain-hc-200M-L18-lr2e3-ladder-20260914
pretrain-hc-200M-L24-lr2e3-ladder-20260914
pretrain-hc-300M-L14-lr2e3-ladder-20260914
pretrain-hc-300M-L21-lr2e3-ladder-20260914
pretrain-hc-300M-L28-lr2e3-ladder-20260914
pretrain-hc-300M-backbone-L110-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L16-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L42-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L56-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L70-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L76-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L80-evenhd-lr2e3-b200-8gpu-final-eval
pretrain-hc-300M-backbone-L84-lr2e3-b200-8gpu-final-eval
pretrain-hc-400M-L16-lr1e3-h100-8gpu-main-final-eval
pretrain-hc-400M-L32-lr1e3-h100-8gpu-main-final-eval
pretrain-hc-500M-L17-lr2e3-ladder-20260914
pretrain-hc-500M-L26-lr2e3-ladder-20260914
pretrain-hc-500M-L34-lr2e3-ladder-20260914
pretrain-llama-130M-hc-s4-lr1e-3
pretrain-llama-130M-hc-s4-lr1e-3-fix1
pretrain-llama-130M-lr1e-3
pretrain-llama-130M-mhc-s4-lr1e-3-mb8-fix1
pretrain-llama-350M-hc-corrected-L16-lr2e3-h100-8gpu-final-eval
pretrain-llama-350M-hc-corrected-L20-lr2e3-h100-8gpu-final-eval
pretrain-llama-350M-hc-corrected-L28-lr2e3-h100-8gpu-final-eval
pretrain-llama-350M-hc-corrected-L32-lr2e3-h100-8gpu-final-eval
pretrain-llama-350M-hc-corrected-l24-lr1e3-h100-4gpu
pretrain-llama-350M-hc-corrected-l24-lr2e3-h100-4gpu
pretrain-llama-350M-hc-corrected-l24-lr5e3-h100-4gpu
pretrain-llama-350M-hc-s4-lr2e-3
pretrain-llama-350M-hc-s4-lr5e-3
pretrain-llama-350M-liger-mhc-linear-l16_h1216_i3248-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l16_h1216_i3248-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l16_h1216_i3248-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l16_h1216_i3248-lr5e4-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l20_h1120_i2992-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l20_h1120_i2992-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l20_h1120_i2992-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l24_h1024_i2736-lr2e3-seed43-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l24_h1024_i2736-lr2e3-seed44-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l26_h992_i2656-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l26_h992_i2656-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l26_h992_i2656-lr2e3-seed43-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l26_h992_i2656-lr2e3-seed44-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l26_h992_i2656-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l28_h960_i2560-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l28_h960_i2560-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l28_h960_i2560-lr2e3-seed43-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l28_h960_i2560-lr2e3-seed44-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l28_h960_i2560-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l30_h928_i2480-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l30_h928_i2480-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l30_h928_i2480-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l32_h896_i2400-lr1e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l32_h896_i2400-lr2e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l32_h896_i2400-lr5e3-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-l32_h896_i2400-lr5e4-h100-4gpu
pretrain-llama-350M-liger-mhc-linear-s4-lr2e-3-h100-4gpu
pretrain-llama-350M-lr2e-3
pretrain-llama-350M-mhc-linear-s4-lr1e-3-b200-4gpu
pretrain-llama-350M-mhc-linear-s4-lr2e-3-b200-4gpu
pretrain-llama-350M-mhc-linear-s4-lr5e-3-h100-4gpu
pretrain-llama-350M-mhc-linear-s4-lr5e-4-h100-4gpu
pretrain-llama-350M-postnorm-baseline-l24-lr1e3
pretrain-llama-350M-postnorm-baseline-l24-lr2e3
pretrain-llama-350M-postnorm-moda-l24-lr1e3
pretrain-llama-350M-postnorm-moda-l24-lr2e3
pretrain-llama-350M-prenorm-moda-l24-lr1e3
pretrain-llama-350M-prenorm-moda-l24-lr2e3
pretrain-llama-350M-prenorm-moda-l24-lr5e3
pretrain-llama-350M-prenorm-moda-l24-lr5e4
pretrain-llama-350M-prenorm-moda-l26_h992_i2656-lr2e3-a100-8gpu
pretrain-llama-350M-prenorm-moda-l28_h960_i2560-lr2e3-a100-8gpu
pretrain-llama-350M-prenorm-moda-l30_h928_i2480-lr2e3-a100-8gpu
pretrain-llama-350M-prenorm-moda-l32_h896_i2400-lr2e3-a100-8gpu
pretrain-llama-400M-hc-L36-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-hc-L42-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-hc-L50-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-hc-L56-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-hc-L62-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-hc-L70-lr2e3-b200-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L36-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L42-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L50-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L56-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L62-lr2e3-h100-8gpu-final-eval
pretrain-llama-400M-liger-mhc-linear-L70-lr2e3-h100-8gpu-final-eval
pretrain-mhc-200M-L12-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-200M-L18-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-200M-L24-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-300M-L14-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-300M-L21-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-300M-L28-lr2e3-hres-gap8-ladder-20260916
pretrain-mhc-300M-backbone-L16-lr2e3-b200-8gpu-final-eval
pretrain-mhc-300M-backbone-L42-lr2e3-b200-8gpu-final-eval
pretrain-mhc-300M-backbone-L56-lr2e3-b200-8gpu-final-eval
pretrain-mhc-500M-L34-lr2e3-hres-gap8-ladder-20260916
Total size
300 GB
Files
10,168
Last updated
Sep 19
Pre-warmed CDN
US EU US EU

Contributors