--- license: llama2 library_name: peft tags: - trl - sft - generated_from_trainer base_model: codellama/CodeLlama-7b-Instruct-hf model-index: - name: Codellama-7b-lora-rps-adapter results: [] --- # Codellama-7b-lora-rps-adapter This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.2971 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 6 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:-----:|:---------------:| | 0.2892 | 3.41 | 13000 | 0.3005 | | 0.2816 | 3.42 | 13050 | 0.2992 | | 0.2564 | 3.44 | 13100 | 0.2995 | | 0.2596 | 3.45 | 13150 | 0.2997 | | 0.3176 | 3.46 | 13200 | 0.2995 | | 0.2509 | 3.48 | 13250 | 0.2992 | | 0.3033 | 3.49 | 13300 | 0.2983 | | 0.2797 | 3.5 | 13350 | 0.2991 | | 0.2555 | 3.52 | 13400 | 0.2989 | | 0.2821 | 3.53 | 13450 | 0.2984 | | 0.2697 | 3.54 | 13500 | 0.2976 | | 0.2505 | 3.56 | 13550 | 0.2982 | | 0.2696 | 3.57 | 13600 | 0.2985 | | 0.2698 | 3.58 | 13650 | 0.2973 | | 0.2966 | 3.59 | 13700 | 0.2981 | | 0.2468 | 3.61 | 13750 | 0.2967 | | 0.2878 | 3.62 | 13800 | 0.2969 | | 0.2876 | 3.63 | 13850 | 0.2984 | | 0.2584 | 3.65 | 13900 | 0.2973 | | 0.2677 | 3.66 | 13950 | 0.2975 | | 0.2923 | 3.67 | 14000 | 0.2969 | | 0.2643 | 3.69 | 14050 | 0.2968 | | 0.2673 | 3.7 | 14100 | 0.2982 | | 0.3016 | 3.71 | 14150 | 0.2967 | | 0.2707 | 3.73 | 14200 | 0.2968 | | 0.2606 | 3.74 | 14250 | 0.2976 | | 0.2957 | 3.75 | 14300 | 0.2963 | | 0.2509 | 3.77 | 14350 | 0.2965 | | 0.2713 | 3.78 | 14400 | 0.2950 | | 0.3041 | 3.79 | 14450 | 0.2968 | | 0.265 | 3.8 | 14500 | 0.2953 | | 0.2726 | 3.82 | 14550 | 0.2955 | | 0.2671 | 3.83 | 14600 | 0.2957 | | 0.2674 | 3.84 | 14650 | 0.2952 | | 0.2516 | 3.86 | 14700 | 0.2959 | | 0.2535 | 3.87 | 14750 | 0.2944 | | 0.2452 | 3.88 | 14800 | 0.2945 | | 0.2733 | 3.9 | 14850 | 0.2945 | | 0.2718 | 3.91 | 14900 | 0.2947 | | 0.257 | 3.92 | 14950 | 0.2939 | | 0.2418 | 3.94 | 15000 | 0.2946 | | 0.2583 | 3.95 | 15050 | 0.2949 | | 0.3093 | 3.96 | 15100 | 0.2939 | | 0.2501 | 3.98 | 15150 | 0.2937 | | 0.2715 | 3.99 | 15200 | 0.2935 | | 0.3209 | 4.0 | 15250 | 0.2930 | | 0.2655 | 4.01 | 15300 | 0.2964 | | 0.2531 | 4.03 | 15350 | 0.2963 | | 0.2417 | 4.04 | 15400 | 0.2967 | | 0.2574 | 4.05 | 15450 | 0.2956 | | 0.2738 | 4.07 | 15500 | 0.2980 | | 0.2488 | 4.08 | 15550 | 0.2967 | | 0.2243 | 4.09 | 15600 | 0.2976 | | 0.2414 | 4.11 | 15650 | 0.2987 | | 0.2318 | 4.12 | 15700 | 0.2978 | | 0.2428 | 4.13 | 15750 | 0.2963 | | 0.2464 | 4.15 | 15800 | 0.3004 | | 0.2717 | 4.16 | 15850 | 0.2971 | | 0.2233 | 4.17 | 15900 | 0.2977 | | 0.2308 | 4.19 | 15950 | 0.2980 | | 0.2387 | 4.2 | 16000 | 0.2967 | | 0.2305 | 4.21 | 16050 | 0.2980 | | 0.2537 | 4.22 | 16100 | 0.2974 | | 0.2893 | 4.24 | 16150 | 0.2974 | | 0.2403 | 4.25 | 16200 | 0.2987 | | 0.218 | 4.26 | 16250 | 0.2970 | | 0.2447 | 4.28 | 16300 | 0.2963 | | 0.2334 | 4.29 | 16350 | 0.2975 | | 0.2481 | 4.3 | 16400 | 0.2980 | | 0.2374 | 4.32 | 16450 | 0.2986 | | 0.2207 | 4.33 | 16500 | 0.2968 | | 0.2746 | 4.34 | 16550 | 0.2968 | | 0.2262 | 4.36 | 16600 | 0.2962 | | 0.2625 | 4.37 | 16650 | 0.2966 | | 0.2946 | 4.38 | 16700 | 0.2969 | | 0.2745 | 4.4 | 16750 | 0.2990 | | 0.2418 | 4.41 | 16800 | 0.2973 | | 0.2501 | 4.42 | 16850 | 0.2964 | | 0.2295 | 4.43 | 16900 | 0.2970 | | 0.2558 | 4.45 | 16950 | 0.2966 | | 0.2577 | 4.46 | 17000 | 0.2967 | | 0.2307 | 4.47 | 17050 | 0.2965 | | 0.2276 | 4.49 | 17100 | 0.2965 | | 0.2574 | 4.5 | 17150 | 0.2963 | | 0.2668 | 4.51 | 17200 | 0.2956 | | 0.2327 | 4.53 | 17250 | 0.2977 | | 0.254 | 4.54 | 17300 | 0.2951 | | 0.2511 | 4.55 | 17350 | 0.2949 | | 0.2176 | 4.57 | 17400 | 0.2953 | | 0.2294 | 4.58 | 17450 | 0.2962 | | 0.2588 | 4.59 | 17500 | 0.2951 | | 0.2903 | 4.61 | 17550 | 0.2946 | | 0.2326 | 4.62 | 17600 | 0.2946 | | 0.2508 | 4.63 | 17650 | 0.2945 | | 0.251 | 4.64 | 17700 | 0.2949 | | 0.2281 | 4.66 | 17750 | 0.2954 | | 0.2265 | 4.67 | 17800 | 0.2958 | | 0.23 | 4.68 | 17850 | 0.2955 | | 0.2489 | 4.7 | 17900 | 0.2946 | | 0.2373 | 4.71 | 17950 | 0.2945 | | 0.2365 | 4.72 | 18000 | 0.2948 | | 0.245 | 4.74 | 18050 | 0.2941 | | 0.2597 | 4.75 | 18100 | 0.2949 | | 0.2325 | 4.76 | 18150 | 0.2946 | | 0.2495 | 4.78 | 18200 | 0.2946 | | 0.2583 | 4.79 | 18250 | 0.2949 | | 0.2162 | 4.8 | 18300 | 0.2945 | | 0.2292 | 4.82 | 18350 | 0.2941 | | 0.2695 | 4.83 | 18400 | 0.2943 | | 0.2602 | 4.84 | 18450 | 0.2934 | | 0.2179 | 4.85 | 18500 | 0.2947 | | 0.2356 | 4.87 | 18550 | 0.2930 | | 0.2692 | 4.88 | 18600 | 0.2936 | | 0.2627 | 4.89 | 18650 | 0.2935 | | 0.2476 | 4.91 | 18700 | 0.2947 | | 0.2529 | 4.92 | 18750 | 0.2924 | | 0.2358 | 4.93 | 18800 | 0.2920 | | 0.232 | 4.95 | 18850 | 0.2930 | | 0.2439 | 4.96 | 18900 | 0.2928 | | 0.2355 | 4.97 | 18950 | 0.2926 | | 0.2489 | 4.99 | 19000 | 0.2926 | | 0.2777 | 5.0 | 19050 | 0.2924 | | 0.2384 | 5.01 | 19100 | 0.2976 | | 0.2201 | 5.02 | 19150 | 0.2990 | | 0.2065 | 5.04 | 19200 | 0.2976 | | 0.2138 | 5.05 | 19250 | 0.2963 | | 0.2367 | 5.06 | 19300 | 0.2973 | | 0.2319 | 5.08 | 19350 | 0.2985 | | 0.2199 | 5.09 | 19400 | 0.2991 | | 0.212 | 5.1 | 19450 | 0.2989 | | 0.2085 | 5.12 | 19500 | 0.2983 | | 0.1995 | 5.13 | 19550 | 0.2986 | | 0.2333 | 5.14 | 19600 | 0.2979 | | 0.2286 | 5.16 | 19650 | 0.2989 | | 0.215 | 5.17 | 19700 | 0.2977 | | 0.2417 | 5.18 | 19750 | 0.2977 | | 0.211 | 5.2 | 19800 | 0.2983 | | 0.2132 | 5.21 | 19850 | 0.2996 | | 0.2079 | 5.22 | 19900 | 0.3001 | | 0.2022 | 5.23 | 19950 | 0.2988 | | 0.2178 | 5.25 | 20000 | 0.2989 | | 0.1975 | 5.26 | 20050 | 0.2992 | | 0.245 | 5.27 | 20100 | 0.2979 | | 0.256 | 5.29 | 20150 | 0.2988 | | 0.2355 | 5.3 | 20200 | 0.3005 | | 0.2018 | 5.31 | 20250 | 0.3001 | | 0.2078 | 5.33 | 20300 | 0.2986 | | 0.2128 | 5.34 | 20350 | 0.2997 | | 0.2202 | 5.35 | 20400 | 0.2979 | | 0.2311 | 5.37 | 20450 | 0.2988 | | 0.2374 | 5.38 | 20500 | 0.2980 | | 0.2617 | 5.39 | 20550 | 0.2975 | | 0.2082 | 5.41 | 20600 | 0.2985 | | 0.2531 | 5.42 | 20650 | 0.2984 | | 0.2103 | 5.43 | 20700 | 0.2979 | | 0.2569 | 5.44 | 20750 | 0.2990 | | 0.2254 | 5.46 | 20800 | 0.2977 | | 0.2328 | 5.47 | 20850 | 0.2971 | | 0.2162 | 5.48 | 20900 | 0.2981 | | 0.2373 | 5.5 | 20950 | 0.2977 | | 0.2447 | 5.51 | 21000 | 0.2972 | | 0.2131 | 5.52 | 21050 | 0.2976 | | 0.2355 | 5.54 | 21100 | 0.2983 | | 0.2327 | 5.55 | 21150 | 0.2977 | | 0.2601 | 5.56 | 21200 | 0.2970 | | 0.2074 | 5.58 | 21250 | 0.2977 | | 0.2153 | 5.59 | 21300 | 0.2969 | | 0.2072 | 5.6 | 21350 | 0.2975 | | 0.2162 | 5.62 | 21400 | 0.2971 | | 0.2399 | 5.63 | 21450 | 0.2969 | | 0.2248 | 5.64 | 21500 | 0.2974 | | 0.2608 | 5.65 | 21550 | 0.2973 | | 0.2322 | 5.67 | 21600 | 0.2974 | | 0.2283 | 5.68 | 21650 | 0.2983 | | 0.1931 | 5.69 | 21700 | 0.2986 | | 0.2341 | 5.71 | 21750 | 0.2981 | | 0.2318 | 5.72 | 21800 | 0.2979 | | 0.223 | 5.73 | 21850 | 0.2974 | | 0.2075 | 5.75 | 21900 | 0.2974 | | 0.2277 | 5.76 | 21950 | 0.2975 | | 0.2167 | 5.77 | 22000 | 0.2976 | | 0.1904 | 5.79 | 22050 | 0.2975 | | 0.2049 | 5.8 | 22100 | 0.2974 | | 0.2207 | 5.81 | 22150 | 0.2976 | | 0.2017 | 5.83 | 22200 | 0.2972 | | 0.2068 | 5.84 | 22250 | 0.2972 | | 0.2453 | 5.85 | 22300 | 0.2967 | | 0.2121 | 5.86 | 22350 | 0.2968 | | 0.2193 | 5.88 | 22400 | 0.2971 | | 0.2337 | 5.89 | 22450 | 0.2971 | | 0.227 | 5.9 | 22500 | 0.2971 | | 0.2285 | 5.92 | 22550 | 0.2972 | | 0.2141 | 5.93 | 22600 | 0.2974 | | 0.2269 | 5.94 | 22650 | 0.2971 | | 0.2192 | 5.96 | 22700 | 0.2971 | | 0.217 | 5.97 | 22750 | 0.2971 | | 0.1998 | 5.98 | 22800 | 0.2971 | | 0.2188 | 6.0 | 22850 | 0.2971 | ### Framework versions - PEFT 0.7.1 - Transformers 4.36.2 - Pytorch 2.1.0+cu121 - Datasets 2.16.1 - Tokenizers 0.15.0