Training in progress, step 10, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
@@ -20,13 +20,13 @@
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
-
"v_proj",
|
24 |
"k_proj",
|
25 |
-
"q_proj",
|
26 |
-
"up_proj",
|
27 |
"gate_proj",
|
|
|
28 |
"down_proj",
|
29 |
-
"
|
|
|
|
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
32 |
"use_dora": false,
|
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
|
|
23 |
"k_proj",
|
|
|
|
|
24 |
"gate_proj",
|
25 |
+
"o_proj",
|
26 |
"down_proj",
|
27 |
+
"v_proj",
|
28 |
+
"up_proj",
|
29 |
+
"q_proj"
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
32 |
"use_dora": false,
|
last-checkpoint/adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 83115256
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:21cb1efe47a7b9ec7582f9c4cad57e9fdc96cddc43b5fe446da11199b86121df
|
3 |
size 83115256
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 166351098
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab11797a5a48c8b5091a20e4ab8445d69bbcf0b52c3045370731dea57405fdcc
|
3 |
size 166351098
|
last-checkpoint/trainer_state.json
CHANGED
@@ -10,7 +10,7 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0005925925925925926,
|
13 |
-
"grad_norm": 10.
|
14 |
"learning_rate": 5e-07,
|
15 |
"logits/chosen": -4.444676399230957,
|
16 |
"logits/rejected": -4.0909342765808105,
|
@@ -25,7 +25,7 @@
|
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.0011851851851851852,
|
28 |
-
"grad_norm": 7.
|
29 |
"learning_rate": 4.849231551964771e-07,
|
30 |
"logits/chosen": -4.530362606048584,
|
31 |
"logits/rejected": -3.981240749359131,
|
@@ -40,122 +40,122 @@
|
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.0017777777777777779,
|
43 |
-
"grad_norm":
|
44 |
"learning_rate": 4.415111107797445e-07,
|
45 |
-
"logits/chosen": -4.
|
46 |
-
"logits/rejected": -4.
|
47 |
-
"logps/chosen": -200.
|
48 |
-
"logps/rejected": -224.
|
49 |
-
"loss": 0.
|
50 |
"rewards/accuracies": 1.0,
|
51 |
-
"rewards/chosen": 0.
|
52 |
-
"rewards/margins": 0.
|
53 |
-
"rewards/rejected": -0.
|
54 |
"step": 3
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.0023703703703703703,
|
58 |
-
"grad_norm":
|
59 |
"learning_rate": 3.75e-07,
|
60 |
-
"logits/chosen": -3.
|
61 |
-
"logits/rejected": -4.
|
62 |
-
"logps/chosen": -239.
|
63 |
-
"logps/rejected": -202.
|
64 |
-
"loss": 0.
|
65 |
"rewards/accuracies": 0.0,
|
66 |
-
"rewards/chosen": -0.
|
67 |
-
"rewards/margins": -0.
|
68 |
-
"rewards/rejected": 0.
|
69 |
"step": 4
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.002962962962962963,
|
73 |
-
"grad_norm": 6.
|
74 |
"learning_rate": 2.934120444167326e-07,
|
75 |
-
"logits/chosen": -4.
|
76 |
-
"logits/rejected": -4.
|
77 |
-
"logps/chosen": -169.
|
78 |
-
"logps/rejected": -149.
|
79 |
-
"loss": 0.
|
80 |
-
"rewards/accuracies": 0.
|
81 |
-
"rewards/chosen": 0.
|
82 |
-
"rewards/margins": 0.
|
83 |
-
"rewards/rejected": 0.
|
84 |
"step": 5
|
85 |
},
|
86 |
{
|
87 |
"epoch": 0.0035555555555555557,
|
88 |
-
"grad_norm": 13.
|
89 |
"learning_rate": 2.065879555832674e-07,
|
90 |
-
"logits/chosen": -5.
|
91 |
-
"logits/rejected": -5.
|
92 |
-
"logps/chosen": -
|
93 |
-
"logps/rejected": -202.
|
94 |
-
"loss": 0.
|
95 |
"rewards/accuracies": 0.75,
|
96 |
-
"rewards/chosen": 0.
|
97 |
-
"rewards/margins": 0.
|
98 |
-
"rewards/rejected": 0.
|
99 |
"step": 6
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.004148148148148148,
|
103 |
-
"grad_norm": 6.
|
104 |
"learning_rate": 1.2500000000000005e-07,
|
105 |
-
"logits/chosen": -4.
|
106 |
-
"logits/rejected": -3.
|
107 |
-
"logps/chosen": -176.
|
108 |
-
"logps/rejected": -222.
|
109 |
-
"loss": 0.
|
110 |
-
"rewards/accuracies": 0.
|
111 |
-
"rewards/chosen": 0.
|
112 |
-
"rewards/margins": -0.
|
113 |
-
"rewards/rejected": 0.
|
114 |
"step": 7
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.004740740740740741,
|
118 |
-
"grad_norm": 10.
|
119 |
"learning_rate": 5.848888922025552e-08,
|
120 |
-
"logits/chosen": -4.
|
121 |
-
"logits/rejected": -4.
|
122 |
-
"logps/chosen": -181.
|
123 |
-
"logps/rejected": -196.
|
124 |
-
"loss": 0.
|
125 |
-
"rewards/accuracies": 0.
|
126 |
-
"rewards/chosen":
|
127 |
-
"rewards/margins":
|
128 |
-
"rewards/rejected": -0.
|
129 |
"step": 8
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.005333333333333333,
|
133 |
-
"grad_norm": 9.
|
134 |
"learning_rate": 1.507684480352292e-08,
|
135 |
-
"logits/chosen": -4.
|
136 |
-
"logits/rejected": -4.
|
137 |
-
"logps/chosen": -154.
|
138 |
-
"logps/rejected": -179.
|
139 |
-
"loss": 0.
|
140 |
-
"rewards/accuracies": 0.
|
141 |
-
"rewards/chosen": 0.
|
142 |
-
"rewards/margins": 0.
|
143 |
-
"rewards/rejected": 0.
|
144 |
"step": 9
|
145 |
},
|
146 |
{
|
147 |
"epoch": 0.005925925925925926,
|
148 |
-
"grad_norm": 10.
|
149 |
"learning_rate": 0.0,
|
150 |
-
"logits/chosen": -3.
|
151 |
-
"logits/rejected": -3.
|
152 |
-
"logps/chosen": -
|
153 |
-
"logps/rejected": -254.
|
154 |
-
"loss": 0.
|
155 |
"rewards/accuracies": 0.5,
|
156 |
-
"rewards/chosen": -0.
|
157 |
-
"rewards/margins": -0.
|
158 |
-
"rewards/rejected": -0.
|
159 |
"step": 10
|
160 |
}
|
161 |
],
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0005925925925925926,
|
13 |
+
"grad_norm": 10.896183013916016,
|
14 |
"learning_rate": 5e-07,
|
15 |
"logits/chosen": -4.444676399230957,
|
16 |
"logits/rejected": -4.0909342765808105,
|
|
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.0011851851851851852,
|
28 |
+
"grad_norm": 7.562353610992432,
|
29 |
"learning_rate": 4.849231551964771e-07,
|
30 |
"logits/chosen": -4.530362606048584,
|
31 |
"logits/rejected": -3.981240749359131,
|
|
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.0017777777777777779,
|
43 |
+
"grad_norm": 13.934762001037598,
|
44 |
"learning_rate": 4.415111107797445e-07,
|
45 |
+
"logits/chosen": -4.541503429412842,
|
46 |
+
"logits/rejected": -4.324827194213867,
|
47 |
+
"logps/chosen": -200.9736328125,
|
48 |
+
"logps/rejected": -224.6981658935547,
|
49 |
+
"loss": 0.6691,
|
50 |
"rewards/accuracies": 1.0,
|
51 |
+
"rewards/chosen": 0.022507095709443092,
|
52 |
+
"rewards/margins": 0.04895630106329918,
|
53 |
+
"rewards/rejected": -0.026449203491210938,
|
54 |
"step": 3
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.0023703703703703703,
|
58 |
+
"grad_norm": 13.249699592590332,
|
59 |
"learning_rate": 3.75e-07,
|
60 |
+
"logits/chosen": -3.9642434120178223,
|
61 |
+
"logits/rejected": -4.461881160736084,
|
62 |
+
"logps/chosen": -239.5960693359375,
|
63 |
+
"logps/rejected": -202.10887145996094,
|
64 |
+
"loss": 0.7259,
|
65 |
"rewards/accuracies": 0.0,
|
66 |
+
"rewards/chosen": -0.001570509746670723,
|
67 |
+
"rewards/margins": -0.06432266533374786,
|
68 |
+
"rewards/rejected": 0.06275215744972229,
|
69 |
"step": 4
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.002962962962962963,
|
73 |
+
"grad_norm": 6.773025989532471,
|
74 |
"learning_rate": 2.934120444167326e-07,
|
75 |
+
"logits/chosen": -4.136728286743164,
|
76 |
+
"logits/rejected": -4.700671672821045,
|
77 |
+
"logps/chosen": -169.1282958984375,
|
78 |
+
"logps/rejected": -149.28570556640625,
|
79 |
+
"loss": 0.6866,
|
80 |
+
"rewards/accuracies": 0.75,
|
81 |
+
"rewards/chosen": 0.017114639282226562,
|
82 |
+
"rewards/margins": 0.013147544115781784,
|
83 |
+
"rewards/rejected": 0.003967094700783491,
|
84 |
"step": 5
|
85 |
},
|
86 |
{
|
87 |
"epoch": 0.0035555555555555557,
|
88 |
+
"grad_norm": 13.418476104736328,
|
89 |
"learning_rate": 2.065879555832674e-07,
|
90 |
+
"logits/chosen": -5.0562872886657715,
|
91 |
+
"logits/rejected": -5.145793914794922,
|
92 |
+
"logps/chosen": -194.10568237304688,
|
93 |
+
"logps/rejected": -202.68751525878906,
|
94 |
+
"loss": 0.6884,
|
95 |
"rewards/accuracies": 0.75,
|
96 |
+
"rewards/chosen": -0.034047700464725494,
|
97 |
+
"rewards/margins": 0.01024474948644638,
|
98 |
+
"rewards/rejected": -0.044292449951171875,
|
99 |
"step": 6
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.004148148148148148,
|
103 |
+
"grad_norm": 6.973136901855469,
|
104 |
"learning_rate": 1.2500000000000005e-07,
|
105 |
+
"logits/chosen": -4.244935035705566,
|
106 |
+
"logits/rejected": -3.681328058242798,
|
107 |
+
"logps/chosen": -176.23780822753906,
|
108 |
+
"logps/rejected": -222.976318359375,
|
109 |
+
"loss": 0.7073,
|
110 |
+
"rewards/accuracies": 0.5,
|
111 |
+
"rewards/chosen": 0.016452789306640625,
|
112 |
+
"rewards/margins": -0.027773665264248848,
|
113 |
+
"rewards/rejected": 0.044226452708244324,
|
114 |
"step": 7
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.004740740740740741,
|
118 |
+
"grad_norm": 10.38602066040039,
|
119 |
"learning_rate": 5.848888922025552e-08,
|
120 |
+
"logits/chosen": -4.409693241119385,
|
121 |
+
"logits/rejected": -4.41237735748291,
|
122 |
+
"logps/chosen": -181.64944458007812,
|
123 |
+
"logps/rejected": -196.00564575195312,
|
124 |
+
"loss": 0.6896,
|
125 |
+
"rewards/accuracies": 0.5,
|
126 |
+
"rewards/chosen": 0.005864143371582031,
|
127 |
+
"rewards/margins": 0.007416536100208759,
|
128 |
+
"rewards/rejected": -0.001552392728626728,
|
129 |
"step": 8
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.005333333333333333,
|
133 |
+
"grad_norm": 9.169404983520508,
|
134 |
"learning_rate": 1.507684480352292e-08,
|
135 |
+
"logits/chosen": -4.903069972991943,
|
136 |
+
"logits/rejected": -4.5977277755737305,
|
137 |
+
"logps/chosen": -154.88751220703125,
|
138 |
+
"logps/rejected": -179.42715454101562,
|
139 |
+
"loss": 0.695,
|
140 |
+
"rewards/accuracies": 0.5,
|
141 |
+
"rewards/chosen": -0.016914749518036842,
|
142 |
+
"rewards/margins": -0.003633500775322318,
|
143 |
+
"rewards/rejected": -0.013281249441206455,
|
144 |
"step": 9
|
145 |
},
|
146 |
{
|
147 |
"epoch": 0.005925925925925926,
|
148 |
+
"grad_norm": 10.337571144104004,
|
149 |
"learning_rate": 0.0,
|
150 |
+
"logits/chosen": -3.817504644393921,
|
151 |
+
"logits/rejected": -3.697187900543213,
|
152 |
+
"logps/chosen": -233.053466796875,
|
153 |
+
"logps/rejected": -254.43130493164062,
|
154 |
+
"loss": 0.7085,
|
155 |
"rewards/accuracies": 0.5,
|
156 |
+
"rewards/chosen": -0.06154556944966316,
|
157 |
+
"rewards/margins": -0.029002761468291283,
|
158 |
+
"rewards/rejected": -0.03254280239343643,
|
159 |
"step": 10
|
160 |
}
|
161 |
],
|
last-checkpoint/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5944
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:265f72a5d7c9bd4d95478bba01ec7511ef89dd45479699ac483e45fdedd76cef
|
3 |
size 5944
|