-
Add 2000-vocab BPE tokenizer for inference serving
-
Add proper model card: 256d/4L/4H/2KV, vocab=2000, distilled from JuliaFluxGPT
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload final_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload final_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload checkpoint_latest.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload checkpoint_latest.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload checkpoint_latest.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload checkpoint_latest.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
Upload best_model.jld2 (39.7 MB)
-
initial commit