Upload folder using huggingface_hub
Browse files- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- trainer_state.json +50 -50
- training_args.bin +1 -1
adapter_config.json
CHANGED
@@ -10,7 +10,7 @@
|
|
10 |
"layers_pattern": null,
|
11 |
"layers_to_transform": null,
|
12 |
"loftq_config": {},
|
13 |
-
"lora_alpha":
|
14 |
"lora_dropout": 0.1,
|
15 |
"megatron_config": null,
|
16 |
"megatron_core": "megatron.core",
|
@@ -20,8 +20,8 @@
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
-
"
|
24 |
-
"
|
25 |
],
|
26 |
"task_type": "CAUSAL_LM",
|
27 |
"use_dora": false,
|
|
|
10 |
"layers_pattern": null,
|
11 |
"layers_to_transform": null,
|
12 |
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
"lora_dropout": 0.1,
|
15 |
"megatron_config": null,
|
16 |
"megatron_core": "megatron.core",
|
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"q_proj"
|
25 |
],
|
26 |
"task_type": "CAUSAL_LM",
|
27 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 109069176
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1912b4b957d908b8bfc710c3e58558d93617ba78eeee6b44cea5846184e311aa
|
3 |
size 109069176
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 218182458
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f61189ba8e2df9508d8734043f68c2833bde41baa22941571ec4278f299a891b
|
3 |
size 218182458
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1e8d2ce26e6c0f5a47ca9bc5544c0f80a1d4937019d9d7af01320826a96817c0
|
3 |
size 14244
|
trainer_state.json
CHANGED
@@ -10,122 +10,122 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.24154589371980675,
|
13 |
-
"grad_norm": 0.
|
14 |
"learning_rate": 0.000493936638139193,
|
15 |
-
"loss": 1.
|
16 |
"step": 25
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.24154589371980675,
|
20 |
-
"eval_loss":
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second": 3.
|
23 |
-
"eval_steps_per_second": 0.
|
24 |
"step": 25
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.4830917874396135,
|
28 |
-
"grad_norm": 0.
|
29 |
"learning_rate": 0.000463751348237005,
|
30 |
-
"loss": 0.
|
31 |
"step": 50
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.4830917874396135,
|
35 |
-
"eval_loss":
|
36 |
-
"eval_runtime":
|
37 |
-
"eval_samples_per_second": 3.
|
38 |
-
"eval_steps_per_second": 0.
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.7246376811594203,
|
43 |
-
"grad_norm": 0.
|
44 |
"learning_rate": 0.000411248712216741,
|
45 |
-
"loss": 0.
|
46 |
"step": 75
|
47 |
},
|
48 |
{
|
49 |
"epoch": 0.7246376811594203,
|
50 |
-
"eval_loss":
|
51 |
-
"eval_runtime":
|
52 |
-
"eval_samples_per_second": 3.
|
53 |
-
"eval_steps_per_second": 0.
|
54 |
"step": 75
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.966183574879227,
|
58 |
-
"grad_norm": 0.
|
59 |
"learning_rate": 0.00034191042415818,
|
60 |
-
"loss": 0.
|
61 |
"step": 100
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.966183574879227,
|
65 |
-
"eval_loss":
|
66 |
-
"eval_runtime":
|
67 |
-
"eval_samples_per_second": 3.
|
68 |
-
"eval_steps_per_second": 0.
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 1.2077294685990339,
|
73 |
-
"grad_norm": 0.
|
74 |
"learning_rate": 0.00026297595453297494,
|
75 |
-
"loss": 0.
|
76 |
"step": 125
|
77 |
},
|
78 |
{
|
79 |
"epoch": 1.2077294685990339,
|
80 |
-
"eval_loss":
|
81 |
-
"eval_runtime":
|
82 |
-
"eval_samples_per_second": 3.
|
83 |
-
"eval_steps_per_second": 0.
|
84 |
"step": 125
|
85 |
},
|
86 |
{
|
87 |
"epoch": 1.4492753623188406,
|
88 |
-
"grad_norm": 0.
|
89 |
"learning_rate": 0.00018268669172909137,
|
90 |
-
"loss": 0.
|
91 |
"step": 150
|
92 |
},
|
93 |
{
|
94 |
"epoch": 1.4492753623188406,
|
95 |
-
"eval_loss":
|
96 |
-
"eval_runtime":
|
97 |
-
"eval_samples_per_second": 3.
|
98 |
-
"eval_steps_per_second": 0.
|
99 |
"step": 150
|
100 |
},
|
101 |
{
|
102 |
"epoch": 1.6908212560386473,
|
103 |
-
"grad_norm": 0.
|
104 |
"learning_rate": 0.00010942547535123056,
|
105 |
-
"loss": 0.
|
106 |
"step": 175
|
107 |
},
|
108 |
{
|
109 |
"epoch": 1.6908212560386473,
|
110 |
-
"eval_loss":
|
111 |
-
"eval_runtime":
|
112 |
-
"eval_samples_per_second": 3.
|
113 |
-
"eval_steps_per_second": 0.
|
114 |
"step": 175
|
115 |
},
|
116 |
{
|
117 |
"epoch": 1.9323671497584543,
|
118 |
-
"grad_norm": 0.
|
119 |
"learning_rate": 5.0841360885691e-05,
|
120 |
-
"loss": 0.
|
121 |
"step": 200
|
122 |
},
|
123 |
{
|
124 |
"epoch": 1.9323671497584543,
|
125 |
-
"eval_loss":
|
126 |
-
"eval_runtime":
|
127 |
-
"eval_samples_per_second": 3.
|
128 |
-
"eval_steps_per_second": 0.
|
129 |
"step": 200
|
130 |
}
|
131 |
],
|
@@ -133,7 +133,7 @@
|
|
133 |
"max_steps": 250,
|
134 |
"num_input_tokens_seen": 0,
|
135 |
"num_train_epochs": 3,
|
136 |
-
"save_steps":
|
137 |
"stateful_callbacks": {
|
138 |
"TrainerControl": {
|
139 |
"args": {
|
@@ -146,7 +146,7 @@
|
|
146 |
"attributes": {}
|
147 |
}
|
148 |
},
|
149 |
-
"total_flos": 1.
|
150 |
"train_batch_size": 1,
|
151 |
"trial_name": null,
|
152 |
"trial_params": null
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.24154589371980675,
|
13 |
+
"grad_norm": 0.24812498688697815,
|
14 |
"learning_rate": 0.000493936638139193,
|
15 |
+
"loss": 1.1502,
|
16 |
"step": 25
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.24154589371980675,
|
20 |
+
"eval_loss": 0.9789943099021912,
|
21 |
+
"eval_runtime": 52.5341,
|
22 |
+
"eval_samples_per_second": 3.94,
|
23 |
+
"eval_steps_per_second": 0.495,
|
24 |
"step": 25
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.4830917874396135,
|
28 |
+
"grad_norm": 0.35792288184165955,
|
29 |
"learning_rate": 0.000463751348237005,
|
30 |
+
"loss": 0.9241,
|
31 |
"step": 50
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.4830917874396135,
|
35 |
+
"eval_loss": 0.9137127995491028,
|
36 |
+
"eval_runtime": 52.1664,
|
37 |
+
"eval_samples_per_second": 3.968,
|
38 |
+
"eval_steps_per_second": 0.498,
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.7246376811594203,
|
43 |
+
"grad_norm": 0.1500280201435089,
|
44 |
"learning_rate": 0.000411248712216741,
|
45 |
+
"loss": 0.8907,
|
46 |
"step": 75
|
47 |
},
|
48 |
{
|
49 |
"epoch": 0.7246376811594203,
|
50 |
+
"eval_loss": 0.871120274066925,
|
51 |
+
"eval_runtime": 52.6024,
|
52 |
+
"eval_samples_per_second": 3.935,
|
53 |
+
"eval_steps_per_second": 0.494,
|
54 |
"step": 75
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.966183574879227,
|
58 |
+
"grad_norm": 0.17570209503173828,
|
59 |
"learning_rate": 0.00034191042415818,
|
60 |
+
"loss": 0.8684,
|
61 |
"step": 100
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.966183574879227,
|
65 |
+
"eval_loss": 0.8509008288383484,
|
66 |
+
"eval_runtime": 52.4802,
|
67 |
+
"eval_samples_per_second": 3.944,
|
68 |
+
"eval_steps_per_second": 0.495,
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 1.2077294685990339,
|
73 |
+
"grad_norm": 0.12721501290798187,
|
74 |
"learning_rate": 0.00026297595453297494,
|
75 |
+
"loss": 0.8553,
|
76 |
"step": 125
|
77 |
},
|
78 |
{
|
79 |
"epoch": 1.2077294685990339,
|
80 |
+
"eval_loss": 0.8410875201225281,
|
81 |
+
"eval_runtime": 52.6568,
|
82 |
+
"eval_samples_per_second": 3.931,
|
83 |
+
"eval_steps_per_second": 0.494,
|
84 |
"step": 125
|
85 |
},
|
86 |
{
|
87 |
"epoch": 1.4492753623188406,
|
88 |
+
"grad_norm": 0.14220421016216278,
|
89 |
"learning_rate": 0.00018268669172909137,
|
90 |
+
"loss": 0.8527,
|
91 |
"step": 150
|
92 |
},
|
93 |
{
|
94 |
"epoch": 1.4492753623188406,
|
95 |
+
"eval_loss": 0.8425164818763733,
|
96 |
+
"eval_runtime": 52.5225,
|
97 |
+
"eval_samples_per_second": 3.941,
|
98 |
+
"eval_steps_per_second": 0.495,
|
99 |
"step": 150
|
100 |
},
|
101 |
{
|
102 |
"epoch": 1.6908212560386473,
|
103 |
+
"grad_norm": 0.13018099963665009,
|
104 |
"learning_rate": 0.00010942547535123056,
|
105 |
+
"loss": 0.8532,
|
106 |
"step": 175
|
107 |
},
|
108 |
{
|
109 |
"epoch": 1.6908212560386473,
|
110 |
+
"eval_loss": 0.83249831199646,
|
111 |
+
"eval_runtime": 52.6913,
|
112 |
+
"eval_samples_per_second": 3.929,
|
113 |
+
"eval_steps_per_second": 0.493,
|
114 |
"step": 175
|
115 |
},
|
116 |
{
|
117 |
"epoch": 1.9323671497584543,
|
118 |
+
"grad_norm": 0.1186065599322319,
|
119 |
"learning_rate": 5.0841360885691e-05,
|
120 |
+
"loss": 0.8463,
|
121 |
"step": 200
|
122 |
},
|
123 |
{
|
124 |
"epoch": 1.9323671497584543,
|
125 |
+
"eval_loss": 0.8275504112243652,
|
126 |
+
"eval_runtime": 52.8803,
|
127 |
+
"eval_samples_per_second": 3.915,
|
128 |
+
"eval_steps_per_second": 0.492,
|
129 |
"step": 200
|
130 |
}
|
131 |
],
|
|
|
133 |
"max_steps": 250,
|
134 |
"num_input_tokens_seen": 0,
|
135 |
"num_train_epochs": 3,
|
136 |
+
"save_steps": 50,
|
137 |
"stateful_callbacks": {
|
138 |
"TrainerControl": {
|
139 |
"args": {
|
|
|
146 |
"attributes": {}
|
147 |
}
|
148 |
},
|
149 |
+
"total_flos": 1.5515046692683776e+16,
|
150 |
"train_batch_size": 1,
|
151 |
"trial_name": null,
|
152 |
"trial_params": null
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5368
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6dfba6deedf548ad55e99786ef825605d398f2faa387a7055cf0e4a125a246d4
|
3 |
size 5368
|