Commit
•
93ad889
1
Parent(s):
96910e9
Training in progress, step 46000
Browse files- adapter_model.bin +1 -1
- {checkpoint-42000 → checkpoint-45000/adapter_model}/README.md +0 -0
- {checkpoint-42000 → checkpoint-45000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-42000 → checkpoint-45000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-42000/adapter_model → checkpoint-46000}/README.md +0 -0
- {checkpoint-42000/adapter_model → checkpoint-46000}/adapter_config.json +0 -0
- {checkpoint-42000/adapter_model → checkpoint-46000}/adapter_model.bin +1 -1
- {checkpoint-42000 → checkpoint-46000}/optimizer.pt +1 -1
- {checkpoint-42000 → checkpoint-46000}/rng_state.pth +1 -1
- {checkpoint-42000 → checkpoint-46000}/scheduler.pt +1 -1
- {checkpoint-42000 → checkpoint-46000}/trainer_state.json +277 -5
- {checkpoint-42000 → checkpoint-46000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d76b0e007a3405c670c4bf90f86029a3bbd2cf891deb3f05f7059777fa176d
|
| 3 |
size 500897101
|
{checkpoint-42000 → checkpoint-45000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-42000 → checkpoint-45000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-42000 → checkpoint-45000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12e83a97e3fac36a6950c710fcd825a7e2fcd307d5e7c3afa5b431a88b7267d
|
| 3 |
size 500897101
|
{checkpoint-42000/adapter_model → checkpoint-46000}/README.md
RENAMED
|
File without changes
|
{checkpoint-42000/adapter_model → checkpoint-46000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-42000/adapter_model → checkpoint-46000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d76b0e007a3405c670c4bf90f86029a3bbd2cf891deb3f05f7059777fa176d
|
| 3 |
size 500897101
|
{checkpoint-42000 → checkpoint-46000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:165047cac7cab265b645033afa4b8ce42d37ca93752428f2db2a2605a4ccf514
|
| 3 |
size 1001723453
|
{checkpoint-42000 → checkpoint-46000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:762307e219d5dcd959d4b72d256b21463c82f844e4f9c33972f4649eaeb35965
|
| 3 |
size 14575
|
{checkpoint-42000 → checkpoint-46000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c427cdcd60ed56d4ba60659268ae055fed051c2dc525cae02d3094e98336390
|
| 3 |
size 627
|
{checkpoint-42000 → checkpoint-46000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -2862,11 +2862,283 @@
|
|
| 2862 |
"eval_samples_per_second": 0.441,
|
| 2863 |
"eval_steps_per_second": 0.441,
|
| 2864 |
"step": 42000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2865 |
}
|
| 2866 |
],
|
| 2867 |
"max_steps": 80463,
|
| 2868 |
"num_train_epochs": 3,
|
| 2869 |
-
"total_flos": 1.
|
| 2870 |
"trial_name": null,
|
| 2871 |
"trial_params": null
|
| 2872 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5225542187690735,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-46000",
|
| 4 |
+
"epoch": 1.7150740091719174,
|
| 5 |
+
"global_step": 46000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 2862 |
"eval_samples_per_second": 0.441,
|
| 2863 |
"eval_steps_per_second": 0.441,
|
| 2864 |
"step": 42000
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 1.57,
|
| 2868 |
+
"learning_rate": 9.272967731398264e-05,
|
| 2869 |
+
"loss": 0.4789,
|
| 2870 |
+
"step": 42100
|
| 2871 |
+
},
|
| 2872 |
+
{
|
| 2873 |
+
"epoch": 1.57,
|
| 2874 |
+
"learning_rate": 9.234027916915512e-05,
|
| 2875 |
+
"loss": 0.4644,
|
| 2876 |
+
"step": 42200
|
| 2877 |
+
},
|
| 2878 |
+
{
|
| 2879 |
+
"epoch": 1.58,
|
| 2880 |
+
"learning_rate": 9.195099782024136e-05,
|
| 2881 |
+
"loss": 0.4517,
|
| 2882 |
+
"step": 42300
|
| 2883 |
+
},
|
| 2884 |
+
{
|
| 2885 |
+
"epoch": 1.58,
|
| 2886 |
+
"learning_rate": 9.156183920302836e-05,
|
| 2887 |
+
"loss": 0.4768,
|
| 2888 |
+
"step": 42400
|
| 2889 |
+
},
|
| 2890 |
+
{
|
| 2891 |
+
"epoch": 1.58,
|
| 2892 |
+
"learning_rate": 9.117280925143156e-05,
|
| 2893 |
+
"loss": 0.4472,
|
| 2894 |
+
"step": 42500
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 1.59,
|
| 2898 |
+
"learning_rate": 9.078391389740465e-05,
|
| 2899 |
+
"loss": 0.5002,
|
| 2900 |
+
"step": 42600
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 1.59,
|
| 2904 |
+
"learning_rate": 9.039515907084884e-05,
|
| 2905 |
+
"loss": 0.4674,
|
| 2906 |
+
"step": 42700
|
| 2907 |
+
},
|
| 2908 |
+
{
|
| 2909 |
+
"epoch": 1.6,
|
| 2910 |
+
"learning_rate": 9.000655069952262e-05,
|
| 2911 |
+
"loss": 0.4704,
|
| 2912 |
+
"step": 42800
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 1.6,
|
| 2916 |
+
"learning_rate": 8.961809470895141e-05,
|
| 2917 |
+
"loss": 0.4671,
|
| 2918 |
+
"step": 42900
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"epoch": 1.6,
|
| 2922 |
+
"learning_rate": 8.922979702233692e-05,
|
| 2923 |
+
"loss": 0.5007,
|
| 2924 |
+
"step": 43000
|
| 2925 |
+
},
|
| 2926 |
+
{
|
| 2927 |
+
"epoch": 1.6,
|
| 2928 |
+
"eval_loss": 0.5325181484222412,
|
| 2929 |
+
"eval_runtime": 1297.5192,
|
| 2930 |
+
"eval_samples_per_second": 0.418,
|
| 2931 |
+
"eval_steps_per_second": 0.418,
|
| 2932 |
+
"step": 43000
|
| 2933 |
+
},
|
| 2934 |
+
{
|
| 2935 |
+
"epoch": 1.61,
|
| 2936 |
+
"learning_rate": 8.884166356046725e-05,
|
| 2937 |
+
"loss": 0.4767,
|
| 2938 |
+
"step": 43100
|
| 2939 |
+
},
|
| 2940 |
+
{
|
| 2941 |
+
"epoch": 1.61,
|
| 2942 |
+
"learning_rate": 8.845370024162619e-05,
|
| 2943 |
+
"loss": 0.517,
|
| 2944 |
+
"step": 43200
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 1.61,
|
| 2948 |
+
"learning_rate": 8.806591298150332e-05,
|
| 2949 |
+
"loss": 0.4742,
|
| 2950 |
+
"step": 43300
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 1.62,
|
| 2954 |
+
"learning_rate": 8.767830769310362e-05,
|
| 2955 |
+
"loss": 0.4953,
|
| 2956 |
+
"step": 43400
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 1.62,
|
| 2960 |
+
"learning_rate": 8.729089028665733e-05,
|
| 2961 |
+
"loss": 0.473,
|
| 2962 |
+
"step": 43500
|
| 2963 |
+
},
|
| 2964 |
+
{
|
| 2965 |
+
"epoch": 1.63,
|
| 2966 |
+
"learning_rate": 8.690366666952989e-05,
|
| 2967 |
+
"loss": 0.4695,
|
| 2968 |
+
"step": 43600
|
| 2969 |
+
},
|
| 2970 |
+
{
|
| 2971 |
+
"epoch": 1.63,
|
| 2972 |
+
"learning_rate": 8.651664274613183e-05,
|
| 2973 |
+
"loss": 0.4948,
|
| 2974 |
+
"step": 43700
|
| 2975 |
+
},
|
| 2976 |
+
{
|
| 2977 |
+
"epoch": 1.63,
|
| 2978 |
+
"learning_rate": 8.612982441782866e-05,
|
| 2979 |
+
"loss": 0.499,
|
| 2980 |
+
"step": 43800
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 1.64,
|
| 2984 |
+
"learning_rate": 8.574321758285104e-05,
|
| 2985 |
+
"loss": 0.48,
|
| 2986 |
+
"step": 43900
|
| 2987 |
+
},
|
| 2988 |
+
{
|
| 2989 |
+
"epoch": 1.64,
|
| 2990 |
+
"learning_rate": 8.535682813620482e-05,
|
| 2991 |
+
"loss": 0.4672,
|
| 2992 |
+
"step": 44000
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 1.64,
|
| 2996 |
+
"eval_loss": 0.5285405516624451,
|
| 2997 |
+
"eval_runtime": 1272.6033,
|
| 2998 |
+
"eval_samples_per_second": 0.426,
|
| 2999 |
+
"eval_steps_per_second": 0.426,
|
| 3000 |
+
"step": 44000
|
| 3001 |
+
},
|
| 3002 |
+
{
|
| 3003 |
+
"epoch": 1.64,
|
| 3004 |
+
"learning_rate": 8.497066196958097e-05,
|
| 3005 |
+
"loss": 0.4684,
|
| 3006 |
+
"step": 44100
|
| 3007 |
+
},
|
| 3008 |
+
{
|
| 3009 |
+
"epoch": 1.65,
|
| 3010 |
+
"learning_rate": 8.458472497126595e-05,
|
| 3011 |
+
"loss": 0.4923,
|
| 3012 |
+
"step": 44200
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 1.65,
|
| 3016 |
+
"learning_rate": 8.41990230260518e-05,
|
| 3017 |
+
"loss": 0.492,
|
| 3018 |
+
"step": 44300
|
| 3019 |
+
},
|
| 3020 |
+
{
|
| 3021 |
+
"epoch": 1.66,
|
| 3022 |
+
"learning_rate": 8.38135620151465e-05,
|
| 3023 |
+
"loss": 0.4814,
|
| 3024 |
+
"step": 44400
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 1.66,
|
| 3028 |
+
"learning_rate": 8.342834781608424e-05,
|
| 3029 |
+
"loss": 0.4753,
|
| 3030 |
+
"step": 44500
|
| 3031 |
+
},
|
| 3032 |
+
{
|
| 3033 |
+
"epoch": 1.66,
|
| 3034 |
+
"learning_rate": 8.304338630263579e-05,
|
| 3035 |
+
"loss": 0.5032,
|
| 3036 |
+
"step": 44600
|
| 3037 |
+
},
|
| 3038 |
+
{
|
| 3039 |
+
"epoch": 1.67,
|
| 3040 |
+
"learning_rate": 8.265868334471895e-05,
|
| 3041 |
+
"loss": 0.4838,
|
| 3042 |
+
"step": 44700
|
| 3043 |
+
},
|
| 3044 |
+
{
|
| 3045 |
+
"epoch": 1.67,
|
| 3046 |
+
"learning_rate": 8.227424480830907e-05,
|
| 3047 |
+
"loss": 0.4463,
|
| 3048 |
+
"step": 44800
|
| 3049 |
+
},
|
| 3050 |
+
{
|
| 3051 |
+
"epoch": 1.67,
|
| 3052 |
+
"learning_rate": 8.189007655534959e-05,
|
| 3053 |
+
"loss": 0.458,
|
| 3054 |
+
"step": 44900
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 1.68,
|
| 3058 |
+
"learning_rate": 8.150618444366255e-05,
|
| 3059 |
+
"loss": 0.4812,
|
| 3060 |
+
"step": 45000
|
| 3061 |
+
},
|
| 3062 |
+
{
|
| 3063 |
+
"epoch": 1.68,
|
| 3064 |
+
"eval_loss": 0.5250784754753113,
|
| 3065 |
+
"eval_runtime": 1265.3534,
|
| 3066 |
+
"eval_samples_per_second": 0.428,
|
| 3067 |
+
"eval_steps_per_second": 0.428,
|
| 3068 |
+
"step": 45000
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 1.68,
|
| 3072 |
+
"learning_rate": 8.112257432685958e-05,
|
| 3073 |
+
"loss": 0.4743,
|
| 3074 |
+
"step": 45100
|
| 3075 |
+
},
|
| 3076 |
+
{
|
| 3077 |
+
"epoch": 1.69,
|
| 3078 |
+
"learning_rate": 8.073925205425225e-05,
|
| 3079 |
+
"loss": 0.4925,
|
| 3080 |
+
"step": 45200
|
| 3081 |
+
},
|
| 3082 |
+
{
|
| 3083 |
+
"epoch": 1.69,
|
| 3084 |
+
"learning_rate": 8.035622347076312e-05,
|
| 3085 |
+
"loss": 0.4748,
|
| 3086 |
+
"step": 45300
|
| 3087 |
+
},
|
| 3088 |
+
{
|
| 3089 |
+
"epoch": 1.69,
|
| 3090 |
+
"learning_rate": 7.997349441683657e-05,
|
| 3091 |
+
"loss": 0.4745,
|
| 3092 |
+
"step": 45400
|
| 3093 |
+
},
|
| 3094 |
+
{
|
| 3095 |
+
"epoch": 1.7,
|
| 3096 |
+
"learning_rate": 7.959107072834971e-05,
|
| 3097 |
+
"loss": 0.449,
|
| 3098 |
+
"step": 45500
|
| 3099 |
+
},
|
| 3100 |
+
{
|
| 3101 |
+
"epoch": 1.7,
|
| 3102 |
+
"learning_rate": 7.920895823652346e-05,
|
| 3103 |
+
"loss": 0.4753,
|
| 3104 |
+
"step": 45600
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 1.7,
|
| 3108 |
+
"learning_rate": 7.882716276783352e-05,
|
| 3109 |
+
"loss": 0.457,
|
| 3110 |
+
"step": 45700
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 1.71,
|
| 3114 |
+
"learning_rate": 7.844569014392172e-05,
|
| 3115 |
+
"loss": 0.488,
|
| 3116 |
+
"step": 45800
|
| 3117 |
+
},
|
| 3118 |
+
{
|
| 3119 |
+
"epoch": 1.71,
|
| 3120 |
+
"learning_rate": 7.806454618150698e-05,
|
| 3121 |
+
"loss": 0.4782,
|
| 3122 |
+
"step": 45900
|
| 3123 |
+
},
|
| 3124 |
+
{
|
| 3125 |
+
"epoch": 1.72,
|
| 3126 |
+
"learning_rate": 7.768373669229688e-05,
|
| 3127 |
+
"loss": 0.4751,
|
| 3128 |
+
"step": 46000
|
| 3129 |
+
},
|
| 3130 |
+
{
|
| 3131 |
+
"epoch": 1.72,
|
| 3132 |
+
"eval_loss": 0.5225542187690735,
|
| 3133 |
+
"eval_runtime": 1263.3928,
|
| 3134 |
+
"eval_samples_per_second": 0.429,
|
| 3135 |
+
"eval_steps_per_second": 0.429,
|
| 3136 |
+
"step": 46000
|
| 3137 |
}
|
| 3138 |
],
|
| 3139 |
"max_steps": 80463,
|
| 3140 |
"num_train_epochs": 3,
|
| 3141 |
+
"total_flos": 1.289994941944406e+19,
|
| 3142 |
"trial_name": null,
|
| 3143 |
"trial_params": null
|
| 3144 |
}
|
{checkpoint-42000 → checkpoint-46000}/training_args.bin
RENAMED
|
File without changes
|