Initial model upload
Browse files- model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +515 -515
- training_args.bin +1 -1
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 540015464
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:86f5fed9db7f1166fc490c00be5c1f8408ae29ffdfdc06e9366ea03088118108
|
3 |
size 540015464
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1080155450
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fbe4e8d00e51a804cb43110a72e1301d3c481d5a64f33e4bf4fb7d094d046d63
|
3 |
size 1080155450
|
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 sha256:6d7ab3d37b2247ffc32f94eb6f11f1b883d0b67a368ad1f1e1ea8a270b8fbc55
|
3 |
size 14244
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17dc218989896c953db8914a12b113693da0de2b0e964476cd0a24374d2b0a52
|
3 |
size 1064
|
trainer_state.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "bkai-fine-tuned-legal/checkpoint-282",
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 282,
|
7 |
"is_hyper_param_search": false,
|
@@ -10,555 +10,555 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 2.
|
15 |
-
"loss":
|
16 |
-
"step":
|
17 |
},
|
18 |
{
|
19 |
"epoch": 1.0,
|
20 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
21 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
22 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
23 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
24 |
-
"eval_dim_128_cosine_map@100": 0.
|
25 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
26 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
27 |
-
"eval_dim_128_cosine_precision@1": 0.
|
28 |
-
"eval_dim_128_cosine_precision@10": 0.
|
29 |
-
"eval_dim_128_cosine_precision@3": 0.
|
30 |
-
"eval_dim_128_cosine_precision@5": 0.
|
31 |
-
"eval_dim_128_cosine_recall@1": 0.
|
32 |
-
"eval_dim_128_cosine_recall@10": 0.
|
33 |
-
"eval_dim_128_cosine_recall@3": 0.
|
34 |
-
"eval_dim_128_cosine_recall@5": 0.
|
35 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
36 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
37 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
38 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
39 |
-
"eval_dim_256_cosine_map@100": 0.
|
40 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
41 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
42 |
-
"eval_dim_256_cosine_precision@1": 0.
|
43 |
-
"eval_dim_256_cosine_precision@10": 0.
|
44 |
-
"eval_dim_256_cosine_precision@3": 0.
|
45 |
-
"eval_dim_256_cosine_precision@5": 0.
|
46 |
-
"eval_dim_256_cosine_recall@1": 0.
|
47 |
-
"eval_dim_256_cosine_recall@10": 0.
|
48 |
-
"eval_dim_256_cosine_recall@3": 0.
|
49 |
-
"eval_dim_256_cosine_recall@5": 0.
|
50 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
51 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
52 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
53 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
54 |
-
"eval_dim_512_cosine_map@100": 0.
|
55 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
56 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
57 |
-
"eval_dim_512_cosine_precision@1": 0.
|
58 |
-
"eval_dim_512_cosine_precision@10": 0.
|
59 |
-
"eval_dim_512_cosine_precision@3": 0.
|
60 |
-
"eval_dim_512_cosine_precision@5": 0.
|
61 |
-
"eval_dim_512_cosine_recall@1": 0.
|
62 |
-
"eval_dim_512_cosine_recall@10": 0.
|
63 |
-
"eval_dim_512_cosine_recall@3": 0.
|
64 |
-
"eval_dim_512_cosine_recall@5": 0.
|
65 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
66 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
67 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
68 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
69 |
-
"eval_dim_64_cosine_map@100": 0.
|
70 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
71 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
72 |
-
"eval_dim_64_cosine_precision@1": 0.
|
73 |
-
"eval_dim_64_cosine_precision@10": 0.
|
74 |
-
"eval_dim_64_cosine_precision@3": 0.
|
75 |
-
"eval_dim_64_cosine_precision@5": 0.
|
76 |
-
"eval_dim_64_cosine_recall@1": 0.
|
77 |
-
"eval_dim_64_cosine_recall@10": 0.
|
78 |
-
"eval_dim_64_cosine_recall@3": 0.
|
79 |
-
"eval_dim_64_cosine_recall@5": 0.
|
80 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
81 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
82 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
83 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
84 |
-
"eval_dim_768_cosine_map@100": 0.
|
85 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
86 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
87 |
-
"eval_dim_768_cosine_precision@1": 0.
|
88 |
-
"eval_dim_768_cosine_precision@10": 0.
|
89 |
-
"eval_dim_768_cosine_precision@3": 0.
|
90 |
-
"eval_dim_768_cosine_precision@5": 0.
|
91 |
-
"eval_dim_768_cosine_recall@1": 0.
|
92 |
-
"eval_dim_768_cosine_recall@10": 0.
|
93 |
-
"eval_dim_768_cosine_recall@3": 0.
|
94 |
-
"eval_dim_768_cosine_recall@5": 0.
|
95 |
-
"eval_loss": 0.
|
96 |
-
"eval_runtime":
|
97 |
-
"eval_samples_per_second": 1.
|
98 |
-
"eval_sequential_score": 0.
|
99 |
-
"eval_steps_per_second": 0.
|
100 |
-
"step":
|
101 |
},
|
102 |
{
|
103 |
"epoch": 2.0,
|
104 |
-
"grad_norm":
|
105 |
-
"learning_rate": 2.
|
106 |
-
"loss": 0.
|
107 |
-
"step":
|
108 |
},
|
109 |
{
|
110 |
"epoch": 2.0,
|
111 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
112 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
113 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
114 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
115 |
-
"eval_dim_128_cosine_map@100": 0.
|
116 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
117 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
118 |
-
"eval_dim_128_cosine_precision@1": 0.
|
119 |
-
"eval_dim_128_cosine_precision@10": 0.
|
120 |
-
"eval_dim_128_cosine_precision@3": 0.
|
121 |
-
"eval_dim_128_cosine_precision@5": 0.
|
122 |
-
"eval_dim_128_cosine_recall@1": 0.
|
123 |
-
"eval_dim_128_cosine_recall@10": 0.
|
124 |
-
"eval_dim_128_cosine_recall@3": 0.
|
125 |
-
"eval_dim_128_cosine_recall@5": 0.
|
126 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
127 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
128 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
129 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
130 |
-
"eval_dim_256_cosine_map@100": 0.
|
131 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
132 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
133 |
-
"eval_dim_256_cosine_precision@1": 0.
|
134 |
-
"eval_dim_256_cosine_precision@10": 0.
|
135 |
-
"eval_dim_256_cosine_precision@3": 0.
|
136 |
-
"eval_dim_256_cosine_precision@5": 0.
|
137 |
-
"eval_dim_256_cosine_recall@1": 0.
|
138 |
-
"eval_dim_256_cosine_recall@10": 0.
|
139 |
-
"eval_dim_256_cosine_recall@3": 0.
|
140 |
-
"eval_dim_256_cosine_recall@5": 0.
|
141 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
142 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
143 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
144 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
145 |
-
"eval_dim_512_cosine_map@100": 0.
|
146 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
147 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
148 |
-
"eval_dim_512_cosine_precision@1": 0.
|
149 |
-
"eval_dim_512_cosine_precision@10": 0.
|
150 |
-
"eval_dim_512_cosine_precision@3": 0.
|
151 |
-
"eval_dim_512_cosine_precision@5": 0.
|
152 |
-
"eval_dim_512_cosine_recall@1": 0.
|
153 |
-
"eval_dim_512_cosine_recall@10": 0.
|
154 |
-
"eval_dim_512_cosine_recall@3": 0.
|
155 |
-
"eval_dim_512_cosine_recall@5": 0.
|
156 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
157 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
158 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
159 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
160 |
-
"eval_dim_64_cosine_map@100": 0.
|
161 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
162 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
163 |
-
"eval_dim_64_cosine_precision@1": 0.
|
164 |
-
"eval_dim_64_cosine_precision@10": 0.
|
165 |
-
"eval_dim_64_cosine_precision@3": 0.
|
166 |
-
"eval_dim_64_cosine_precision@5": 0.
|
167 |
-
"eval_dim_64_cosine_recall@1": 0.
|
168 |
-
"eval_dim_64_cosine_recall@10": 0.
|
169 |
-
"eval_dim_64_cosine_recall@3": 0.
|
170 |
-
"eval_dim_64_cosine_recall@5": 0.
|
171 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
172 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
173 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
174 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
175 |
-
"eval_dim_768_cosine_map@100": 0.
|
176 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
177 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
178 |
-
"eval_dim_768_cosine_precision@1": 0.
|
179 |
-
"eval_dim_768_cosine_precision@10": 0.
|
180 |
-
"eval_dim_768_cosine_precision@3": 0.
|
181 |
-
"eval_dim_768_cosine_precision@5": 0.
|
182 |
-
"eval_dim_768_cosine_recall@1": 0.
|
183 |
-
"eval_dim_768_cosine_recall@10": 0.
|
184 |
-
"eval_dim_768_cosine_recall@3": 0.
|
185 |
-
"eval_dim_768_cosine_recall@5": 0.
|
186 |
-
"eval_loss": 0.
|
187 |
-
"eval_runtime":
|
188 |
-
"eval_samples_per_second": 1.
|
189 |
-
"eval_sequential_score": 0.
|
190 |
-
"eval_steps_per_second": 0.
|
191 |
-
"step":
|
192 |
},
|
193 |
{
|
194 |
"epoch": 3.0,
|
195 |
-
"grad_norm":
|
196 |
-
"learning_rate":
|
197 |
-
"loss": 0.
|
198 |
-
"step":
|
199 |
},
|
200 |
{
|
201 |
"epoch": 3.0,
|
202 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
203 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
204 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
205 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
206 |
-
"eval_dim_128_cosine_map@100": 0.
|
207 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
208 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
209 |
-
"eval_dim_128_cosine_precision@1": 0.
|
210 |
-
"eval_dim_128_cosine_precision@10": 0.
|
211 |
-
"eval_dim_128_cosine_precision@3": 0.
|
212 |
-
"eval_dim_128_cosine_precision@5": 0.
|
213 |
-
"eval_dim_128_cosine_recall@1": 0.
|
214 |
-
"eval_dim_128_cosine_recall@10": 0.
|
215 |
-
"eval_dim_128_cosine_recall@3": 0.
|
216 |
-
"eval_dim_128_cosine_recall@5": 0.
|
217 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
218 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
219 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
220 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
221 |
-
"eval_dim_256_cosine_map@100": 0.
|
222 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
223 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
224 |
-
"eval_dim_256_cosine_precision@1": 0.
|
225 |
-
"eval_dim_256_cosine_precision@10": 0.
|
226 |
-
"eval_dim_256_cosine_precision@3": 0.
|
227 |
-
"eval_dim_256_cosine_precision@5": 0.
|
228 |
-
"eval_dim_256_cosine_recall@1": 0.
|
229 |
-
"eval_dim_256_cosine_recall@10": 0.
|
230 |
-
"eval_dim_256_cosine_recall@3": 0.
|
231 |
-
"eval_dim_256_cosine_recall@5": 0.
|
232 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
233 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
234 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
235 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
236 |
-
"eval_dim_512_cosine_map@100": 0.
|
237 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
238 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
239 |
-
"eval_dim_512_cosine_precision@1": 0.
|
240 |
-
"eval_dim_512_cosine_precision@10": 0.
|
241 |
-
"eval_dim_512_cosine_precision@3": 0.
|
242 |
-
"eval_dim_512_cosine_precision@5": 0.
|
243 |
-
"eval_dim_512_cosine_recall@1": 0.
|
244 |
-
"eval_dim_512_cosine_recall@10": 0.
|
245 |
-
"eval_dim_512_cosine_recall@3": 0.
|
246 |
-
"eval_dim_512_cosine_recall@5": 0.
|
247 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
248 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
249 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
250 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
251 |
-
"eval_dim_64_cosine_map@100": 0.
|
252 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
253 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
254 |
-
"eval_dim_64_cosine_precision@1": 0.
|
255 |
-
"eval_dim_64_cosine_precision@10": 0.
|
256 |
-
"eval_dim_64_cosine_precision@3": 0.
|
257 |
-
"eval_dim_64_cosine_precision@5": 0.
|
258 |
-
"eval_dim_64_cosine_recall@1": 0.
|
259 |
-
"eval_dim_64_cosine_recall@10": 0.
|
260 |
-
"eval_dim_64_cosine_recall@3": 0.
|
261 |
-
"eval_dim_64_cosine_recall@5": 0.
|
262 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
263 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
264 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
265 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
266 |
-
"eval_dim_768_cosine_map@100": 0.
|
267 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
268 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
269 |
-
"eval_dim_768_cosine_precision@1": 0.
|
270 |
-
"eval_dim_768_cosine_precision@10": 0.
|
271 |
-
"eval_dim_768_cosine_precision@3": 0.
|
272 |
-
"eval_dim_768_cosine_precision@5": 0.
|
273 |
-
"eval_dim_768_cosine_recall@1": 0.
|
274 |
-
"eval_dim_768_cosine_recall@10": 0.
|
275 |
-
"eval_dim_768_cosine_recall@3": 0.
|
276 |
-
"eval_dim_768_cosine_recall@5": 0.
|
277 |
-
"eval_loss": 0.
|
278 |
-
"eval_runtime":
|
279 |
-
"eval_samples_per_second": 1.
|
280 |
-
"eval_sequential_score": 0.
|
281 |
-
"eval_steps_per_second": 0.
|
282 |
-
"step":
|
283 |
},
|
284 |
{
|
285 |
"epoch": 4.0,
|
286 |
-
"grad_norm":
|
287 |
-
"learning_rate":
|
288 |
-
"loss": 0.
|
289 |
-
"step":
|
290 |
},
|
291 |
{
|
292 |
"epoch": 4.0,
|
293 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
294 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
295 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
296 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
297 |
-
"eval_dim_128_cosine_map@100": 0.
|
298 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
299 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
300 |
-
"eval_dim_128_cosine_precision@1": 0.
|
301 |
-
"eval_dim_128_cosine_precision@10": 0.
|
302 |
-
"eval_dim_128_cosine_precision@3": 0.
|
303 |
-
"eval_dim_128_cosine_precision@5": 0.
|
304 |
-
"eval_dim_128_cosine_recall@1": 0.
|
305 |
-
"eval_dim_128_cosine_recall@10": 0.
|
306 |
-
"eval_dim_128_cosine_recall@3": 0.
|
307 |
-
"eval_dim_128_cosine_recall@5": 0.
|
308 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
309 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
310 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
311 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
312 |
-
"eval_dim_256_cosine_map@100": 0.
|
313 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
314 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
315 |
-
"eval_dim_256_cosine_precision@1": 0.
|
316 |
-
"eval_dim_256_cosine_precision@10": 0.
|
317 |
-
"eval_dim_256_cosine_precision@3": 0.
|
318 |
-
"eval_dim_256_cosine_precision@5": 0.
|
319 |
-
"eval_dim_256_cosine_recall@1": 0.
|
320 |
-
"eval_dim_256_cosine_recall@10": 0.
|
321 |
-
"eval_dim_256_cosine_recall@3": 0.
|
322 |
-
"eval_dim_256_cosine_recall@5": 0.
|
323 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
324 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
325 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
326 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
327 |
-
"eval_dim_512_cosine_map@100": 0.
|
328 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
329 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
330 |
-
"eval_dim_512_cosine_precision@1": 0.
|
331 |
-
"eval_dim_512_cosine_precision@10": 0.
|
332 |
-
"eval_dim_512_cosine_precision@3": 0.
|
333 |
-
"eval_dim_512_cosine_precision@5": 0.
|
334 |
-
"eval_dim_512_cosine_recall@1": 0.
|
335 |
-
"eval_dim_512_cosine_recall@10": 0.
|
336 |
-
"eval_dim_512_cosine_recall@3": 0.
|
337 |
-
"eval_dim_512_cosine_recall@5": 0.
|
338 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
339 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
340 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
341 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
342 |
-
"eval_dim_64_cosine_map@100": 0.
|
343 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
344 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
345 |
-
"eval_dim_64_cosine_precision@1": 0.
|
346 |
-
"eval_dim_64_cosine_precision@10": 0.
|
347 |
-
"eval_dim_64_cosine_precision@3": 0.
|
348 |
-
"eval_dim_64_cosine_precision@5": 0.
|
349 |
-
"eval_dim_64_cosine_recall@1": 0.
|
350 |
-
"eval_dim_64_cosine_recall@10": 0.
|
351 |
-
"eval_dim_64_cosine_recall@3": 0.
|
352 |
-
"eval_dim_64_cosine_recall@5": 0.
|
353 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
354 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
355 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
356 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
357 |
-
"eval_dim_768_cosine_map@100": 0.
|
358 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
359 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
360 |
-
"eval_dim_768_cosine_precision@1": 0.
|
361 |
-
"eval_dim_768_cosine_precision@10": 0.
|
362 |
-
"eval_dim_768_cosine_precision@3": 0.
|
363 |
-
"eval_dim_768_cosine_precision@5": 0.
|
364 |
-
"eval_dim_768_cosine_recall@1": 0.
|
365 |
-
"eval_dim_768_cosine_recall@10": 0.
|
366 |
-
"eval_dim_768_cosine_recall@3": 0.
|
367 |
-
"eval_dim_768_cosine_recall@5": 0.
|
368 |
-
"eval_loss": 0.
|
369 |
-
"eval_runtime":
|
370 |
-
"eval_samples_per_second": 1.
|
371 |
-
"eval_sequential_score": 0.
|
372 |
-
"eval_steps_per_second": 0.
|
373 |
-
"step":
|
374 |
},
|
375 |
{
|
376 |
"epoch": 5.0,
|
377 |
-
"grad_norm":
|
378 |
-
"learning_rate":
|
379 |
-
"loss": 0.
|
380 |
-
"step":
|
381 |
},
|
382 |
{
|
383 |
"epoch": 5.0,
|
384 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
385 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
386 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
387 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
388 |
-
"eval_dim_128_cosine_map@100": 0.
|
389 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
390 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
391 |
-
"eval_dim_128_cosine_precision@1": 0.
|
392 |
-
"eval_dim_128_cosine_precision@10": 0.
|
393 |
-
"eval_dim_128_cosine_precision@3": 0.
|
394 |
-
"eval_dim_128_cosine_precision@5": 0.
|
395 |
-
"eval_dim_128_cosine_recall@1": 0.
|
396 |
-
"eval_dim_128_cosine_recall@10": 0.
|
397 |
-
"eval_dim_128_cosine_recall@3": 0.
|
398 |
-
"eval_dim_128_cosine_recall@5": 0.
|
399 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
400 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
401 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
402 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
403 |
-
"eval_dim_256_cosine_map@100": 0.
|
404 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
405 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
406 |
-
"eval_dim_256_cosine_precision@1": 0.
|
407 |
-
"eval_dim_256_cosine_precision@10": 0.
|
408 |
-
"eval_dim_256_cosine_precision@3": 0.
|
409 |
-
"eval_dim_256_cosine_precision@5": 0.
|
410 |
-
"eval_dim_256_cosine_recall@1": 0.
|
411 |
-
"eval_dim_256_cosine_recall@10": 0.
|
412 |
-
"eval_dim_256_cosine_recall@3": 0.
|
413 |
-
"eval_dim_256_cosine_recall@5": 0.
|
414 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
415 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
416 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
417 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
418 |
-
"eval_dim_512_cosine_map@100": 0.
|
419 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
420 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
421 |
-
"eval_dim_512_cosine_precision@1": 0.
|
422 |
-
"eval_dim_512_cosine_precision@10": 0.
|
423 |
-
"eval_dim_512_cosine_precision@3": 0.
|
424 |
-
"eval_dim_512_cosine_precision@5": 0.
|
425 |
-
"eval_dim_512_cosine_recall@1": 0.
|
426 |
-
"eval_dim_512_cosine_recall@10": 0.
|
427 |
-
"eval_dim_512_cosine_recall@3": 0.
|
428 |
-
"eval_dim_512_cosine_recall@5": 0.
|
429 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
430 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
431 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
432 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
433 |
-
"eval_dim_64_cosine_map@100": 0.
|
434 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
435 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
436 |
-
"eval_dim_64_cosine_precision@1": 0.
|
437 |
-
"eval_dim_64_cosine_precision@10": 0.
|
438 |
-
"eval_dim_64_cosine_precision@3": 0.
|
439 |
-
"eval_dim_64_cosine_precision@5": 0.
|
440 |
-
"eval_dim_64_cosine_recall@1": 0.
|
441 |
-
"eval_dim_64_cosine_recall@10": 0.
|
442 |
-
"eval_dim_64_cosine_recall@3": 0.
|
443 |
-
"eval_dim_64_cosine_recall@5": 0.
|
444 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
445 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
446 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
447 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
448 |
-
"eval_dim_768_cosine_map@100": 0.
|
449 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
450 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
451 |
-
"eval_dim_768_cosine_precision@1": 0.
|
452 |
-
"eval_dim_768_cosine_precision@10": 0.
|
453 |
-
"eval_dim_768_cosine_precision@3": 0.
|
454 |
-
"eval_dim_768_cosine_precision@5": 0.
|
455 |
-
"eval_dim_768_cosine_recall@1": 0.
|
456 |
-
"eval_dim_768_cosine_recall@10": 0.
|
457 |
-
"eval_dim_768_cosine_recall@3": 0.
|
458 |
-
"eval_dim_768_cosine_recall@5": 0.
|
459 |
-
"eval_loss": 0.
|
460 |
-
"eval_runtime":
|
461 |
-
"eval_samples_per_second": 1.
|
462 |
-
"eval_sequential_score": 0.
|
463 |
-
"eval_steps_per_second": 0.
|
464 |
-
"step":
|
465 |
},
|
466 |
{
|
467 |
-
"epoch":
|
468 |
-
"grad_norm":
|
469 |
-
"learning_rate":
|
470 |
-
"loss": 0.
|
471 |
"step": 282
|
472 |
},
|
473 |
{
|
474 |
-
"epoch":
|
475 |
-
"eval_dim_128_cosine_accuracy@1": 0.
|
476 |
-
"eval_dim_128_cosine_accuracy@10": 0.
|
477 |
-
"eval_dim_128_cosine_accuracy@3": 0.
|
478 |
-
"eval_dim_128_cosine_accuracy@5": 0.
|
479 |
-
"eval_dim_128_cosine_map@100": 0.
|
480 |
-
"eval_dim_128_cosine_mrr@10": 0.
|
481 |
-
"eval_dim_128_cosine_ndcg@10": 0.
|
482 |
-
"eval_dim_128_cosine_precision@1": 0.
|
483 |
-
"eval_dim_128_cosine_precision@10": 0.
|
484 |
-
"eval_dim_128_cosine_precision@3": 0.
|
485 |
-
"eval_dim_128_cosine_precision@5": 0.
|
486 |
-
"eval_dim_128_cosine_recall@1": 0.
|
487 |
-
"eval_dim_128_cosine_recall@10": 0.
|
488 |
-
"eval_dim_128_cosine_recall@3": 0.
|
489 |
-
"eval_dim_128_cosine_recall@5": 0.
|
490 |
-
"eval_dim_256_cosine_accuracy@1": 0.
|
491 |
-
"eval_dim_256_cosine_accuracy@10": 0.
|
492 |
-
"eval_dim_256_cosine_accuracy@3": 0.
|
493 |
-
"eval_dim_256_cosine_accuracy@5": 0.
|
494 |
-
"eval_dim_256_cosine_map@100": 0.
|
495 |
-
"eval_dim_256_cosine_mrr@10": 0.
|
496 |
-
"eval_dim_256_cosine_ndcg@10": 0.
|
497 |
-
"eval_dim_256_cosine_precision@1": 0.
|
498 |
-
"eval_dim_256_cosine_precision@10": 0.
|
499 |
-
"eval_dim_256_cosine_precision@3": 0.
|
500 |
-
"eval_dim_256_cosine_precision@5": 0.
|
501 |
-
"eval_dim_256_cosine_recall@1": 0.
|
502 |
-
"eval_dim_256_cosine_recall@10": 0.
|
503 |
-
"eval_dim_256_cosine_recall@3": 0.
|
504 |
-
"eval_dim_256_cosine_recall@5": 0.
|
505 |
-
"eval_dim_512_cosine_accuracy@1": 0.
|
506 |
-
"eval_dim_512_cosine_accuracy@10": 0.
|
507 |
-
"eval_dim_512_cosine_accuracy@3": 0.
|
508 |
-
"eval_dim_512_cosine_accuracy@5": 0.
|
509 |
-
"eval_dim_512_cosine_map@100": 0.
|
510 |
-
"eval_dim_512_cosine_mrr@10": 0.
|
511 |
-
"eval_dim_512_cosine_ndcg@10": 0.
|
512 |
-
"eval_dim_512_cosine_precision@1": 0.
|
513 |
-
"eval_dim_512_cosine_precision@10": 0.
|
514 |
-
"eval_dim_512_cosine_precision@3": 0.
|
515 |
-
"eval_dim_512_cosine_precision@5": 0.
|
516 |
-
"eval_dim_512_cosine_recall@1": 0.
|
517 |
-
"eval_dim_512_cosine_recall@10": 0.
|
518 |
-
"eval_dim_512_cosine_recall@3": 0.
|
519 |
-
"eval_dim_512_cosine_recall@5": 0.
|
520 |
-
"eval_dim_64_cosine_accuracy@1": 0.
|
521 |
-
"eval_dim_64_cosine_accuracy@10": 0.
|
522 |
-
"eval_dim_64_cosine_accuracy@3": 0.
|
523 |
-
"eval_dim_64_cosine_accuracy@5": 0.
|
524 |
-
"eval_dim_64_cosine_map@100": 0.
|
525 |
-
"eval_dim_64_cosine_mrr@10": 0.
|
526 |
-
"eval_dim_64_cosine_ndcg@10": 0.
|
527 |
-
"eval_dim_64_cosine_precision@1": 0.
|
528 |
-
"eval_dim_64_cosine_precision@10": 0.
|
529 |
-
"eval_dim_64_cosine_precision@3": 0.
|
530 |
-
"eval_dim_64_cosine_precision@5": 0.
|
531 |
-
"eval_dim_64_cosine_recall@1": 0.
|
532 |
-
"eval_dim_64_cosine_recall@10": 0.
|
533 |
-
"eval_dim_64_cosine_recall@3": 0.
|
534 |
-
"eval_dim_64_cosine_recall@5": 0.
|
535 |
-
"eval_dim_768_cosine_accuracy@1": 0.
|
536 |
-
"eval_dim_768_cosine_accuracy@10": 0.
|
537 |
-
"eval_dim_768_cosine_accuracy@3": 0.
|
538 |
-
"eval_dim_768_cosine_accuracy@5": 0.
|
539 |
-
"eval_dim_768_cosine_map@100": 0.
|
540 |
-
"eval_dim_768_cosine_mrr@10": 0.
|
541 |
-
"eval_dim_768_cosine_ndcg@10": 0.
|
542 |
-
"eval_dim_768_cosine_precision@1": 0.
|
543 |
-
"eval_dim_768_cosine_precision@10": 0.
|
544 |
-
"eval_dim_768_cosine_precision@3": 0.
|
545 |
-
"eval_dim_768_cosine_precision@5": 0.
|
546 |
-
"eval_dim_768_cosine_recall@1": 0.
|
547 |
-
"eval_dim_768_cosine_recall@10": 0.
|
548 |
-
"eval_dim_768_cosine_recall@3": 0.
|
549 |
-
"eval_dim_768_cosine_recall@5": 0.
|
550 |
-
"eval_loss": 0.
|
551 |
-
"eval_runtime":
|
552 |
-
"eval_samples_per_second": 1.
|
553 |
-
"eval_sequential_score": 0.
|
554 |
-
"eval_steps_per_second": 0.
|
555 |
"step": 282
|
556 |
}
|
557 |
],
|
558 |
"logging_steps": 500,
|
559 |
-
"max_steps":
|
560 |
"num_input_tokens_seen": 0,
|
561 |
-
"num_train_epochs":
|
562 |
"save_steps": 500,
|
563 |
"stateful_callbacks": {
|
564 |
"EarlyStoppingCallback": {
|
@@ -582,7 +582,7 @@
|
|
582 |
}
|
583 |
},
|
584 |
"total_flos": 0.0,
|
585 |
-
"train_batch_size":
|
586 |
"trial_name": null,
|
587 |
"trial_params": null
|
588 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.1151694729924202,
|
3 |
"best_model_checkpoint": "bkai-fine-tuned-legal/checkpoint-282",
|
4 |
+
"epoch": 6.0,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 282,
|
7 |
"is_hyper_param_search": false,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"grad_norm": 13545922.0,
|
14 |
+
"learning_rate": 2.877551020408163e-05,
|
15 |
+
"loss": 0.9848,
|
16 |
+
"step": 47
|
17 |
},
|
18 |
{
|
19 |
"epoch": 1.0,
|
20 |
+
"eval_dim_128_cosine_accuracy@1": 0.22911497105045492,
|
21 |
+
"eval_dim_128_cosine_accuracy@10": 0.6825199889716018,
|
22 |
+
"eval_dim_128_cosine_accuracy@3": 0.4164598842018197,
|
23 |
+
"eval_dim_128_cosine_accuracy@5": 0.5108905431486077,
|
24 |
+
"eval_dim_128_cosine_map@100": 0.36449560290881217,
|
25 |
+
"eval_dim_128_cosine_mrr@10": 0.3540410654657966,
|
26 |
+
"eval_dim_128_cosine_ndcg@10": 0.43115214388174655,
|
27 |
+
"eval_dim_128_cosine_precision@1": 0.22911497105045492,
|
28 |
+
"eval_dim_128_cosine_precision@10": 0.06825199889716019,
|
29 |
+
"eval_dim_128_cosine_precision@3": 0.13881996140060654,
|
30 |
+
"eval_dim_128_cosine_precision@5": 0.10217810862972154,
|
31 |
+
"eval_dim_128_cosine_recall@1": 0.22911497105045492,
|
32 |
+
"eval_dim_128_cosine_recall@10": 0.6825199889716018,
|
33 |
+
"eval_dim_128_cosine_recall@3": 0.4164598842018197,
|
34 |
+
"eval_dim_128_cosine_recall@5": 0.5108905431486077,
|
35 |
+
"eval_dim_256_cosine_accuracy@1": 0.2321477805348773,
|
36 |
+
"eval_dim_256_cosine_accuracy@10": 0.6916184174248691,
|
37 |
+
"eval_dim_256_cosine_accuracy@3": 0.4230769230769231,
|
38 |
+
"eval_dim_256_cosine_accuracy@5": 0.5130962227736421,
|
39 |
+
"eval_dim_256_cosine_map@100": 0.3690942352859619,
|
40 |
+
"eval_dim_256_cosine_mrr@10": 0.3585785182559389,
|
41 |
+
"eval_dim_256_cosine_ndcg@10": 0.4367532861198457,
|
42 |
+
"eval_dim_256_cosine_precision@1": 0.2321477805348773,
|
43 |
+
"eval_dim_256_cosine_precision@10": 0.06916184174248691,
|
44 |
+
"eval_dim_256_cosine_precision@3": 0.14102564102564102,
|
45 |
+
"eval_dim_256_cosine_precision@5": 0.10261924455472843,
|
46 |
+
"eval_dim_256_cosine_recall@1": 0.2321477805348773,
|
47 |
+
"eval_dim_256_cosine_recall@10": 0.6916184174248691,
|
48 |
+
"eval_dim_256_cosine_recall@3": 0.4230769230769231,
|
49 |
+
"eval_dim_256_cosine_recall@5": 0.5130962227736421,
|
50 |
+
"eval_dim_512_cosine_accuracy@1": 0.23917838433967467,
|
51 |
+
"eval_dim_512_cosine_accuracy@10": 0.69520264681555,
|
52 |
+
"eval_dim_512_cosine_accuracy@3": 0.43134822167080233,
|
53 |
+
"eval_dim_512_cosine_accuracy@5": 0.5199889716018748,
|
54 |
+
"eval_dim_512_cosine_map@100": 0.3745175599603196,
|
55 |
+
"eval_dim_512_cosine_mrr@10": 0.36392991715572354,
|
56 |
+
"eval_dim_512_cosine_ndcg@10": 0.4416805421730885,
|
57 |
+
"eval_dim_512_cosine_precision@1": 0.23917838433967467,
|
58 |
+
"eval_dim_512_cosine_precision@10": 0.06952026468155499,
|
59 |
+
"eval_dim_512_cosine_precision@3": 0.1437827405569341,
|
60 |
+
"eval_dim_512_cosine_precision@5": 0.10399779432037497,
|
61 |
+
"eval_dim_512_cosine_recall@1": 0.23917838433967467,
|
62 |
+
"eval_dim_512_cosine_recall@10": 0.69520264681555,
|
63 |
+
"eval_dim_512_cosine_recall@3": 0.43134822167080233,
|
64 |
+
"eval_dim_512_cosine_recall@5": 0.5199889716018748,
|
65 |
+
"eval_dim_64_cosine_accuracy@1": 0.21560518334711884,
|
66 |
+
"eval_dim_64_cosine_accuracy@10": 0.6567411083540116,
|
67 |
+
"eval_dim_64_cosine_accuracy@3": 0.3977116073890267,
|
68 |
+
"eval_dim_64_cosine_accuracy@5": 0.4885580369451337,
|
69 |
+
"eval_dim_64_cosine_map@100": 0.34758882194744606,
|
70 |
+
"eval_dim_64_cosine_mrr@10": 0.33651789270068805,
|
71 |
+
"eval_dim_64_cosine_ndcg@10": 0.4116663918808296,
|
72 |
+
"eval_dim_64_cosine_precision@1": 0.21560518334711884,
|
73 |
+
"eval_dim_64_cosine_precision@10": 0.06567411083540116,
|
74 |
+
"eval_dim_64_cosine_precision@3": 0.13257053579634223,
|
75 |
+
"eval_dim_64_cosine_precision@5": 0.09771160738902675,
|
76 |
+
"eval_dim_64_cosine_recall@1": 0.21560518334711884,
|
77 |
+
"eval_dim_64_cosine_recall@10": 0.6567411083540116,
|
78 |
+
"eval_dim_64_cosine_recall@3": 0.3977116073890267,
|
79 |
+
"eval_dim_64_cosine_recall@5": 0.4885580369451337,
|
80 |
+
"eval_dim_768_cosine_accuracy@1": 0.23959194926936864,
|
81 |
+
"eval_dim_768_cosine_accuracy@10": 0.6958919216983733,
|
82 |
+
"eval_dim_768_cosine_accuracy@3": 0.4251447477253929,
|
83 |
+
"eval_dim_768_cosine_accuracy@5": 0.5180590019299697,
|
84 |
+
"eval_dim_768_cosine_map@100": 0.3741732451540211,
|
85 |
+
"eval_dim_768_cosine_mrr@10": 0.3636057391433741,
|
86 |
+
"eval_dim_768_cosine_ndcg@10": 0.4415248560932574,
|
87 |
+
"eval_dim_768_cosine_precision@1": 0.23959194926936864,
|
88 |
+
"eval_dim_768_cosine_precision@10": 0.06958919216983735,
|
89 |
+
"eval_dim_768_cosine_precision@3": 0.1417149159084643,
|
90 |
+
"eval_dim_768_cosine_precision@5": 0.10361180038599395,
|
91 |
+
"eval_dim_768_cosine_recall@1": 0.23959194926936864,
|
92 |
+
"eval_dim_768_cosine_recall@10": 0.6958919216983733,
|
93 |
+
"eval_dim_768_cosine_recall@3": 0.4251447477253929,
|
94 |
+
"eval_dim_768_cosine_recall@5": 0.5180590019299697,
|
95 |
+
"eval_loss": 0.2813453674316406,
|
96 |
+
"eval_runtime": 4767.8829,
|
97 |
+
"eval_samples_per_second": 1.521,
|
98 |
+
"eval_sequential_score": 0.4116663918808296,
|
99 |
+
"eval_steps_per_second": 0.015,
|
100 |
+
"step": 47
|
101 |
},
|
102 |
{
|
103 |
"epoch": 2.0,
|
104 |
+
"grad_norm": 1141820.875,
|
105 |
+
"learning_rate": 2.80333195308751e-05,
|
106 |
+
"loss": 0.2581,
|
107 |
+
"step": 94
|
108 |
},
|
109 |
{
|
110 |
"epoch": 2.0,
|
111 |
+
"eval_dim_128_cosine_accuracy@1": 0.34642955610697546,
|
112 |
+
"eval_dim_128_cosine_accuracy@10": 0.710366694237662,
|
113 |
+
"eval_dim_128_cosine_accuracy@3": 0.4665012406947891,
|
114 |
+
"eval_dim_128_cosine_accuracy@5": 0.5431486076647367,
|
115 |
+
"eval_dim_128_cosine_map@100": 0.4475983290216904,
|
116 |
+
"eval_dim_128_cosine_mrr@10": 0.43726198134800415,
|
117 |
+
"eval_dim_128_cosine_ndcg@10": 0.5006042152520018,
|
118 |
+
"eval_dim_128_cosine_precision@1": 0.34642955610697546,
|
119 |
+
"eval_dim_128_cosine_precision@10": 0.07103666942376619,
|
120 |
+
"eval_dim_128_cosine_precision@3": 0.15550041356492966,
|
121 |
+
"eval_dim_128_cosine_precision@5": 0.10862972153294734,
|
122 |
+
"eval_dim_128_cosine_recall@1": 0.34642955610697546,
|
123 |
+
"eval_dim_128_cosine_recall@10": 0.710366694237662,
|
124 |
+
"eval_dim_128_cosine_recall@3": 0.4665012406947891,
|
125 |
+
"eval_dim_128_cosine_recall@5": 0.5431486076647367,
|
126 |
+
"eval_dim_256_cosine_accuracy@1": 0.3544251447477254,
|
127 |
+
"eval_dim_256_cosine_accuracy@10": 0.7185001378549766,
|
128 |
+
"eval_dim_256_cosine_accuracy@3": 0.47325613454645715,
|
129 |
+
"eval_dim_256_cosine_accuracy@5": 0.5559691204852495,
|
130 |
+
"eval_dim_256_cosine_map@100": 0.45584390323135104,
|
131 |
+
"eval_dim_256_cosine_mrr@10": 0.445744493459548,
|
132 |
+
"eval_dim_256_cosine_ndcg@10": 0.5090183585380493,
|
133 |
+
"eval_dim_256_cosine_precision@1": 0.3544251447477254,
|
134 |
+
"eval_dim_256_cosine_precision@10": 0.07185001378549766,
|
135 |
+
"eval_dim_256_cosine_precision@3": 0.15775204484881902,
|
136 |
+
"eval_dim_256_cosine_precision@5": 0.1111938240970499,
|
137 |
+
"eval_dim_256_cosine_recall@1": 0.3544251447477254,
|
138 |
+
"eval_dim_256_cosine_recall@10": 0.7185001378549766,
|
139 |
+
"eval_dim_256_cosine_recall@3": 0.47325613454645715,
|
140 |
+
"eval_dim_256_cosine_recall@5": 0.5559691204852495,
|
141 |
+
"eval_dim_512_cosine_accuracy@1": 0.3567686793493245,
|
142 |
+
"eval_dim_512_cosine_accuracy@10": 0.7150537634408602,
|
143 |
+
"eval_dim_512_cosine_accuracy@3": 0.47173972980424594,
|
144 |
+
"eval_dim_512_cosine_accuracy@5": 0.5483870967741935,
|
145 |
+
"eval_dim_512_cosine_map@100": 0.4556986968605821,
|
146 |
+
"eval_dim_512_cosine_mrr@10": 0.44540330239255094,
|
147 |
+
"eval_dim_512_cosine_ndcg@10": 0.5078518813826275,
|
148 |
+
"eval_dim_512_cosine_precision@1": 0.3567686793493245,
|
149 |
+
"eval_dim_512_cosine_precision@10": 0.07150537634408603,
|
150 |
+
"eval_dim_512_cosine_precision@3": 0.15724657660141528,
|
151 |
+
"eval_dim_512_cosine_precision@5": 0.10967741935483873,
|
152 |
+
"eval_dim_512_cosine_recall@1": 0.3567686793493245,
|
153 |
+
"eval_dim_512_cosine_recall@10": 0.7150537634408602,
|
154 |
+
"eval_dim_512_cosine_recall@3": 0.47173972980424594,
|
155 |
+
"eval_dim_512_cosine_recall@5": 0.5483870967741935,
|
156 |
+
"eval_dim_64_cosine_accuracy@1": 0.33319547835676866,
|
157 |
+
"eval_dim_64_cosine_accuracy@10": 0.6909291425420457,
|
158 |
+
"eval_dim_64_cosine_accuracy@3": 0.45561069754618144,
|
159 |
+
"eval_dim_64_cosine_accuracy@5": 0.5264681555004136,
|
160 |
+
"eval_dim_64_cosine_map@100": 0.43420213794982415,
|
161 |
+
"eval_dim_64_cosine_mrr@10": 0.423261987912527,
|
162 |
+
"eval_dim_64_cosine_ndcg@10": 0.485318042924584,
|
163 |
+
"eval_dim_64_cosine_precision@1": 0.33319547835676866,
|
164 |
+
"eval_dim_64_cosine_precision@10": 0.06909291425420458,
|
165 |
+
"eval_dim_64_cosine_precision@3": 0.15187023251539378,
|
166 |
+
"eval_dim_64_cosine_precision@5": 0.10529363110008272,
|
167 |
+
"eval_dim_64_cosine_recall@1": 0.33319547835676866,
|
168 |
+
"eval_dim_64_cosine_recall@10": 0.6909291425420457,
|
169 |
+
"eval_dim_64_cosine_recall@3": 0.45561069754618144,
|
170 |
+
"eval_dim_64_cosine_recall@5": 0.5264681555004136,
|
171 |
+
"eval_dim_768_cosine_accuracy@1": 0.3573200992555831,
|
172 |
+
"eval_dim_768_cosine_accuracy@10": 0.7186379928315412,
|
173 |
+
"eval_dim_768_cosine_accuracy@3": 0.47463468431210365,
|
174 |
+
"eval_dim_768_cosine_accuracy@5": 0.5536255858836504,
|
175 |
+
"eval_dim_768_cosine_map@100": 0.45721772728143356,
|
176 |
+
"eval_dim_768_cosine_mrr@10": 0.44721303189045225,
|
177 |
+
"eval_dim_768_cosine_ndcg@10": 0.5101217185995918,
|
178 |
+
"eval_dim_768_cosine_precision@1": 0.3573200992555831,
|
179 |
+
"eval_dim_768_cosine_precision@10": 0.07186379928315412,
|
180 |
+
"eval_dim_768_cosine_precision@3": 0.15821156143736786,
|
181 |
+
"eval_dim_768_cosine_precision@5": 0.11072511717673009,
|
182 |
+
"eval_dim_768_cosine_recall@1": 0.3573200992555831,
|
183 |
+
"eval_dim_768_cosine_recall@10": 0.7186379928315412,
|
184 |
+
"eval_dim_768_cosine_recall@3": 0.47463468431210365,
|
185 |
+
"eval_dim_768_cosine_recall@5": 0.5536255858836504,
|
186 |
+
"eval_loss": 0.14884237945079803,
|
187 |
+
"eval_runtime": 4759.7308,
|
188 |
+
"eval_samples_per_second": 1.524,
|
189 |
+
"eval_sequential_score": 0.485318042924584,
|
190 |
+
"eval_steps_per_second": 0.015,
|
191 |
+
"step": 94
|
192 |
},
|
193 |
{
|
194 |
"epoch": 3.0,
|
195 |
+
"grad_norm": 631988.5625,
|
196 |
+
"learning_rate": 2.2350017747386573e-05,
|
197 |
+
"loss": 0.1103,
|
198 |
+
"step": 141
|
199 |
},
|
200 |
{
|
201 |
"epoch": 3.0,
|
202 |
+
"eval_dim_128_cosine_accuracy@1": 0.34518886131789356,
|
203 |
+
"eval_dim_128_cosine_accuracy@10": 0.7136752136752137,
|
204 |
+
"eval_dim_128_cosine_accuracy@3": 0.4652605459057072,
|
205 |
+
"eval_dim_128_cosine_accuracy@5": 0.5453542872897712,
|
206 |
+
"eval_dim_128_cosine_map@100": 0.4475263336497484,
|
207 |
+
"eval_dim_128_cosine_mrr@10": 0.43706887497210256,
|
208 |
+
"eval_dim_128_cosine_ndcg@10": 0.5011834586328874,
|
209 |
+
"eval_dim_128_cosine_precision@1": 0.34518886131789356,
|
210 |
+
"eval_dim_128_cosine_precision@10": 0.07136752136752138,
|
211 |
+
"eval_dim_128_cosine_precision@3": 0.15508684863523572,
|
212 |
+
"eval_dim_128_cosine_precision@5": 0.10907085745795424,
|
213 |
+
"eval_dim_128_cosine_recall@1": 0.34518886131789356,
|
214 |
+
"eval_dim_128_cosine_recall@10": 0.7136752136752137,
|
215 |
+
"eval_dim_128_cosine_recall@3": 0.4652605459057072,
|
216 |
+
"eval_dim_128_cosine_recall@5": 0.5453542872897712,
|
217 |
+
"eval_dim_256_cosine_accuracy@1": 0.35663082437275984,
|
218 |
+
"eval_dim_256_cosine_accuracy@10": 0.721119382409705,
|
219 |
+
"eval_dim_256_cosine_accuracy@3": 0.4747725392886683,
|
220 |
+
"eval_dim_256_cosine_accuracy@5": 0.5532120209539564,
|
221 |
+
"eval_dim_256_cosine_map@100": 0.4574183781576292,
|
222 |
+
"eval_dim_256_cosine_mrr@10": 0.44702468479350255,
|
223 |
+
"eval_dim_256_cosine_ndcg@10": 0.510536629040665,
|
224 |
+
"eval_dim_256_cosine_precision@1": 0.35663082437275984,
|
225 |
+
"eval_dim_256_cosine_precision@10": 0.07211193824097051,
|
226 |
+
"eval_dim_256_cosine_precision@3": 0.15825751309622277,
|
227 |
+
"eval_dim_256_cosine_precision@5": 0.11064240419079127,
|
228 |
+
"eval_dim_256_cosine_recall@1": 0.35663082437275984,
|
229 |
+
"eval_dim_256_cosine_recall@10": 0.721119382409705,
|
230 |
+
"eval_dim_256_cosine_recall@3": 0.4747725392886683,
|
231 |
+
"eval_dim_256_cosine_recall@5": 0.5532120209539564,
|
232 |
+
"eval_dim_512_cosine_accuracy@1": 0.35842293906810035,
|
233 |
+
"eval_dim_512_cosine_accuracy@10": 0.721119382409705,
|
234 |
+
"eval_dim_512_cosine_accuracy@3": 0.47559966914805624,
|
235 |
+
"eval_dim_512_cosine_accuracy@5": 0.5563826854149435,
|
236 |
+
"eval_dim_512_cosine_map@100": 0.45974376489260765,
|
237 |
+
"eval_dim_512_cosine_mrr@10": 0.4494139522365346,
|
238 |
+
"eval_dim_512_cosine_ndcg@10": 0.5124359611383629,
|
239 |
+
"eval_dim_512_cosine_precision@1": 0.35842293906810035,
|
240 |
+
"eval_dim_512_cosine_precision@10": 0.07211193824097051,
|
241 |
+
"eval_dim_512_cosine_precision@3": 0.15853322304935208,
|
242 |
+
"eval_dim_512_cosine_precision@5": 0.11127653708298871,
|
243 |
+
"eval_dim_512_cosine_recall@1": 0.35842293906810035,
|
244 |
+
"eval_dim_512_cosine_recall@10": 0.721119382409705,
|
245 |
+
"eval_dim_512_cosine_recall@3": 0.47559966914805624,
|
246 |
+
"eval_dim_512_cosine_recall@5": 0.5563826854149435,
|
247 |
+
"eval_dim_64_cosine_accuracy@1": 0.34105321202095396,
|
248 |
+
"eval_dim_64_cosine_accuracy@10": 0.7065067548938516,
|
249 |
+
"eval_dim_64_cosine_accuracy@3": 0.4608491866556383,
|
250 |
+
"eval_dim_64_cosine_accuracy@5": 0.5311552247036118,
|
251 |
+
"eval_dim_64_cosine_map@100": 0.44168965737654264,
|
252 |
+
"eval_dim_64_cosine_mrr@10": 0.4313846000673967,
|
253 |
+
"eval_dim_64_cosine_ndcg@10": 0.4950181594630296,
|
254 |
+
"eval_dim_64_cosine_precision@1": 0.34105321202095396,
|
255 |
+
"eval_dim_64_cosine_precision@10": 0.07065067548938517,
|
256 |
+
"eval_dim_64_cosine_precision@3": 0.15361639555187942,
|
257 |
+
"eval_dim_64_cosine_precision@5": 0.10623104494072236,
|
258 |
+
"eval_dim_64_cosine_recall@1": 0.34105321202095396,
|
259 |
+
"eval_dim_64_cosine_recall@10": 0.7065067548938516,
|
260 |
+
"eval_dim_64_cosine_recall@3": 0.4608491866556383,
|
261 |
+
"eval_dim_64_cosine_recall@5": 0.5311552247036118,
|
262 |
+
"eval_dim_768_cosine_accuracy@1": 0.3617314585056521,
|
263 |
+
"eval_dim_768_cosine_accuracy@10": 0.7263578715191619,
|
264 |
+
"eval_dim_768_cosine_accuracy@3": 0.4779432037496554,
|
265 |
+
"eval_dim_768_cosine_accuracy@5": 0.5548662806727322,
|
266 |
+
"eval_dim_768_cosine_map@100": 0.46162170882821363,
|
267 |
+
"eval_dim_768_cosine_mrr@10": 0.45147318830114674,
|
268 |
+
"eval_dim_768_cosine_ndcg@10": 0.5150788839244959,
|
269 |
+
"eval_dim_768_cosine_precision@1": 0.3617314585056521,
|
270 |
+
"eval_dim_768_cosine_precision@10": 0.0726357871519162,
|
271 |
+
"eval_dim_768_cosine_precision@3": 0.1593144012498851,
|
272 |
+
"eval_dim_768_cosine_precision@5": 0.11097325613454646,
|
273 |
+
"eval_dim_768_cosine_recall@1": 0.3617314585056521,
|
274 |
+
"eval_dim_768_cosine_recall@10": 0.7263578715191619,
|
275 |
+
"eval_dim_768_cosine_recall@3": 0.4779432037496554,
|
276 |
+
"eval_dim_768_cosine_recall@5": 0.5548662806727322,
|
277 |
+
"eval_loss": 0.1296069622039795,
|
278 |
+
"eval_runtime": 4760.1319,
|
279 |
+
"eval_samples_per_second": 1.524,
|
280 |
+
"eval_sequential_score": 0.4950181594630296,
|
281 |
+
"eval_steps_per_second": 0.015,
|
282 |
+
"step": 141
|
283 |
},
|
284 |
{
|
285 |
"epoch": 4.0,
|
286 |
+
"grad_norm": 300185.0,
|
287 |
+
"learning_rate": 1.4568522073177993e-05,
|
288 |
+
"loss": 0.0606,
|
289 |
+
"step": 188
|
290 |
},
|
291 |
{
|
292 |
"epoch": 4.0,
|
293 |
+
"eval_dim_128_cosine_accuracy@1": 0.35497656465398403,
|
294 |
+
"eval_dim_128_cosine_accuracy@10": 0.7209815274331404,
|
295 |
+
"eval_dim_128_cosine_accuracy@3": 0.4738075544527157,
|
296 |
+
"eval_dim_128_cosine_accuracy@5": 0.5551419906258616,
|
297 |
+
"eval_dim_128_cosine_map@100": 0.45665439605706615,
|
298 |
+
"eval_dim_128_cosine_mrr@10": 0.4465567437341632,
|
299 |
+
"eval_dim_128_cosine_ndcg@10": 0.5101629970534793,
|
300 |
+
"eval_dim_128_cosine_precision@1": 0.35497656465398403,
|
301 |
+
"eval_dim_128_cosine_precision@10": 0.07209815274331403,
|
302 |
+
"eval_dim_128_cosine_precision@3": 0.15793585148423855,
|
303 |
+
"eval_dim_128_cosine_precision@5": 0.11102839812517233,
|
304 |
+
"eval_dim_128_cosine_recall@1": 0.35497656465398403,
|
305 |
+
"eval_dim_128_cosine_recall@10": 0.7209815274331404,
|
306 |
+
"eval_dim_128_cosine_recall@3": 0.4738075544527157,
|
307 |
+
"eval_dim_128_cosine_recall@5": 0.5551419906258616,
|
308 |
+
"eval_dim_256_cosine_accuracy@1": 0.35690653432588915,
|
309 |
+
"eval_dim_256_cosine_accuracy@10": 0.7244279018472567,
|
310 |
+
"eval_dim_256_cosine_accuracy@3": 0.4731182795698925,
|
311 |
+
"eval_dim_256_cosine_accuracy@5": 0.5523848910945686,
|
312 |
+
"eval_dim_256_cosine_map@100": 0.4573413953666118,
|
313 |
+
"eval_dim_256_cosine_mrr@10": 0.44724891794784466,
|
314 |
+
"eval_dim_256_cosine_ndcg@10": 0.5113804990740188,
|
315 |
+
"eval_dim_256_cosine_precision@1": 0.35690653432588915,
|
316 |
+
"eval_dim_256_cosine_precision@10": 0.07244279018472567,
|
317 |
+
"eval_dim_256_cosine_precision@3": 0.15770609318996415,
|
318 |
+
"eval_dim_256_cosine_precision@5": 0.1104769782189137,
|
319 |
+
"eval_dim_256_cosine_recall@1": 0.35690653432588915,
|
320 |
+
"eval_dim_256_cosine_recall@10": 0.7244279018472567,
|
321 |
+
"eval_dim_256_cosine_recall@3": 0.4731182795698925,
|
322 |
+
"eval_dim_256_cosine_recall@5": 0.5523848910945686,
|
323 |
+
"eval_dim_512_cosine_accuracy@1": 0.358974358974359,
|
324 |
+
"eval_dim_512_cosine_accuracy@10": 0.7220843672456576,
|
325 |
+
"eval_dim_512_cosine_accuracy@3": 0.47559966914805624,
|
326 |
+
"eval_dim_512_cosine_accuracy@5": 0.5500413564929694,
|
327 |
+
"eval_dim_512_cosine_map@100": 0.4589573057461163,
|
328 |
+
"eval_dim_512_cosine_mrr@10": 0.448363464492498,
|
329 |
+
"eval_dim_512_cosine_ndcg@10": 0.5116861942509243,
|
330 |
+
"eval_dim_512_cosine_precision@1": 0.358974358974359,
|
331 |
+
"eval_dim_512_cosine_precision@10": 0.07220843672456577,
|
332 |
+
"eval_dim_512_cosine_precision@3": 0.15853322304935208,
|
333 |
+
"eval_dim_512_cosine_precision@5": 0.11000827129859388,
|
334 |
+
"eval_dim_512_cosine_recall@1": 0.358974358974359,
|
335 |
+
"eval_dim_512_cosine_recall@10": 0.7220843672456576,
|
336 |
+
"eval_dim_512_cosine_recall@3": 0.47559966914805624,
|
337 |
+
"eval_dim_512_cosine_recall@5": 0.5500413564929694,
|
338 |
+
"eval_dim_64_cosine_accuracy@1": 0.3436724565756824,
|
339 |
+
"eval_dim_64_cosine_accuracy@10": 0.7070581748001102,
|
340 |
+
"eval_dim_64_cosine_accuracy@3": 0.4627791563275434,
|
341 |
+
"eval_dim_64_cosine_accuracy@5": 0.5394265232974911,
|
342 |
+
"eval_dim_64_cosine_map@100": 0.44458140696370635,
|
343 |
+
"eval_dim_64_cosine_mrr@10": 0.43377425044091816,
|
344 |
+
"eval_dim_64_cosine_ndcg@10": 0.49702952074450985,
|
345 |
+
"eval_dim_64_cosine_precision@1": 0.3436724565756824,
|
346 |
+
"eval_dim_64_cosine_precision@10": 0.07070581748001104,
|
347 |
+
"eval_dim_64_cosine_precision@3": 0.15425971877584782,
|
348 |
+
"eval_dim_64_cosine_precision@5": 0.1078853046594982,
|
349 |
+
"eval_dim_64_cosine_recall@1": 0.3436724565756824,
|
350 |
+
"eval_dim_64_cosine_recall@10": 0.7070581748001102,
|
351 |
+
"eval_dim_64_cosine_recall@3": 0.4627791563275434,
|
352 |
+
"eval_dim_64_cosine_recall@5": 0.5394265232974911,
|
353 |
+
"eval_dim_768_cosine_accuracy@1": 0.36200716845878134,
|
354 |
+
"eval_dim_768_cosine_accuracy@10": 0.7274607113316791,
|
355 |
+
"eval_dim_768_cosine_accuracy@3": 0.48249241797628895,
|
356 |
+
"eval_dim_768_cosine_accuracy@5": 0.5601047697821891,
|
357 |
+
"eval_dim_768_cosine_map@100": 0.4636649639575912,
|
358 |
+
"eval_dim_768_cosine_mrr@10": 0.45356874805799724,
|
359 |
+
"eval_dim_768_cosine_ndcg@10": 0.5170505403807711,
|
360 |
+
"eval_dim_768_cosine_precision@1": 0.36200716845878134,
|
361 |
+
"eval_dim_768_cosine_precision@10": 0.07274607113316792,
|
362 |
+
"eval_dim_768_cosine_precision@3": 0.1608308059920963,
|
363 |
+
"eval_dim_768_cosine_precision@5": 0.11202095395643781,
|
364 |
+
"eval_dim_768_cosine_recall@1": 0.36200716845878134,
|
365 |
+
"eval_dim_768_cosine_recall@10": 0.7274607113316791,
|
366 |
+
"eval_dim_768_cosine_recall@3": 0.48249241797628895,
|
367 |
+
"eval_dim_768_cosine_recall@5": 0.5601047697821891,
|
368 |
+
"eval_loss": 0.12068094313144684,
|
369 |
+
"eval_runtime": 4758.824,
|
370 |
+
"eval_samples_per_second": 1.524,
|
371 |
+
"eval_sequential_score": 0.49702952074450985,
|
372 |
+
"eval_steps_per_second": 0.015,
|
373 |
+
"step": 188
|
374 |
},
|
375 |
{
|
376 |
"epoch": 5.0,
|
377 |
+
"grad_norm": 123932.1328125,
|
378 |
+
"learning_rate": 6.9101994825120144e-06,
|
379 |
+
"loss": 0.0413,
|
380 |
+
"step": 235
|
381 |
},
|
382 |
{
|
383 |
"epoch": 5.0,
|
384 |
+
"eval_dim_128_cosine_accuracy@1": 0.35001378549765644,
|
385 |
+
"eval_dim_128_cosine_accuracy@10": 0.7169837331127654,
|
386 |
+
"eval_dim_128_cosine_accuracy@3": 0.4718775847808106,
|
387 |
+
"eval_dim_128_cosine_accuracy@5": 0.5479735318444996,
|
388 |
+
"eval_dim_128_cosine_map@100": 0.4524911878455968,
|
389 |
+
"eval_dim_128_cosine_mrr@10": 0.4418932521083066,
|
390 |
+
"eval_dim_128_cosine_ndcg@10": 0.5056909760387599,
|
391 |
+
"eval_dim_128_cosine_precision@1": 0.35001378549765644,
|
392 |
+
"eval_dim_128_cosine_precision@10": 0.07169837331127654,
|
393 |
+
"eval_dim_128_cosine_precision@3": 0.1572925282602702,
|
394 |
+
"eval_dim_128_cosine_precision@5": 0.10959470636889991,
|
395 |
+
"eval_dim_128_cosine_recall@1": 0.35001378549765644,
|
396 |
+
"eval_dim_128_cosine_recall@10": 0.7169837331127654,
|
397 |
+
"eval_dim_128_cosine_recall@3": 0.4718775847808106,
|
398 |
+
"eval_dim_128_cosine_recall@5": 0.5479735318444996,
|
399 |
+
"eval_dim_256_cosine_accuracy@1": 0.3551144196305487,
|
400 |
+
"eval_dim_256_cosine_accuracy@10": 0.7259443065894678,
|
401 |
+
"eval_dim_256_cosine_accuracy@3": 0.474496829335539,
|
402 |
+
"eval_dim_256_cosine_accuracy@5": 0.5562448304383788,
|
403 |
+
"eval_dim_256_cosine_map@100": 0.45777798616812915,
|
404 |
+
"eval_dim_256_cosine_mrr@10": 0.44761773471451,
|
405 |
+
"eval_dim_256_cosine_ndcg@10": 0.5121486751554244,
|
406 |
+
"eval_dim_256_cosine_precision@1": 0.3551144196305487,
|
407 |
+
"eval_dim_256_cosine_precision@10": 0.07259443065894679,
|
408 |
+
"eval_dim_256_cosine_precision@3": 0.158165609778513,
|
409 |
+
"eval_dim_256_cosine_precision@5": 0.11124896608767576,
|
410 |
+
"eval_dim_256_cosine_recall@1": 0.3551144196305487,
|
411 |
+
"eval_dim_256_cosine_recall@10": 0.7259443065894678,
|
412 |
+
"eval_dim_256_cosine_recall@3": 0.474496829335539,
|
413 |
+
"eval_dim_256_cosine_recall@5": 0.5562448304383788,
|
414 |
+
"eval_dim_512_cosine_accuracy@1": 0.35745795423214777,
|
415 |
+
"eval_dim_512_cosine_accuracy@10": 0.724703611800386,
|
416 |
+
"eval_dim_512_cosine_accuracy@3": 0.4762889440308795,
|
417 |
+
"eval_dim_512_cosine_accuracy@5": 0.5519713261648745,
|
418 |
+
"eval_dim_512_cosine_map@100": 0.45831810667590944,
|
419 |
+
"eval_dim_512_cosine_mrr@10": 0.4480486956293418,
|
420 |
+
"eval_dim_512_cosine_ndcg@10": 0.512088304480363,
|
421 |
+
"eval_dim_512_cosine_precision@1": 0.35745795423214777,
|
422 |
+
"eval_dim_512_cosine_precision@10": 0.0724703611800386,
|
423 |
+
"eval_dim_512_cosine_precision@3": 0.15876298134362649,
|
424 |
+
"eval_dim_512_cosine_precision@5": 0.11039426523297491,
|
425 |
+
"eval_dim_512_cosine_recall@1": 0.35745795423214777,
|
426 |
+
"eval_dim_512_cosine_recall@10": 0.724703611800386,
|
427 |
+
"eval_dim_512_cosine_recall@3": 0.4762889440308795,
|
428 |
+
"eval_dim_512_cosine_recall@5": 0.5519713261648745,
|
429 |
+
"eval_dim_64_cosine_accuracy@1": 0.3403639371381307,
|
430 |
+
"eval_dim_64_cosine_accuracy@10": 0.710366694237662,
|
431 |
+
"eval_dim_64_cosine_accuracy@3": 0.4583677970774745,
|
432 |
+
"eval_dim_64_cosine_accuracy@5": 0.5363937138130687,
|
433 |
+
"eval_dim_64_cosine_map@100": 0.4420078079211827,
|
434 |
+
"eval_dim_64_cosine_mrr@10": 0.4315533083006209,
|
435 |
+
"eval_dim_64_cosine_ndcg@10": 0.49610418689723323,
|
436 |
+
"eval_dim_64_cosine_precision@1": 0.3403639371381307,
|
437 |
+
"eval_dim_64_cosine_precision@10": 0.07103666942376621,
|
438 |
+
"eval_dim_64_cosine_precision@3": 0.1527892656924915,
|
439 |
+
"eval_dim_64_cosine_precision@5": 0.10727874276261373,
|
440 |
+
"eval_dim_64_cosine_recall@1": 0.3403639371381307,
|
441 |
+
"eval_dim_64_cosine_recall@10": 0.710366694237662,
|
442 |
+
"eval_dim_64_cosine_recall@3": 0.4583677970774745,
|
443 |
+
"eval_dim_64_cosine_recall@5": 0.5363937138130687,
|
444 |
+
"eval_dim_768_cosine_accuracy@1": 0.3602150537634409,
|
445 |
+
"eval_dim_768_cosine_accuracy@10": 0.7300799558864075,
|
446 |
+
"eval_dim_768_cosine_accuracy@3": 0.48345740281224153,
|
447 |
+
"eval_dim_768_cosine_accuracy@5": 0.5621725944306589,
|
448 |
+
"eval_dim_768_cosine_map@100": 0.4634583310653756,
|
449 |
+
"eval_dim_768_cosine_mrr@10": 0.4532543621253309,
|
450 |
+
"eval_dim_768_cosine_ndcg@10": 0.517503271654276,
|
451 |
+
"eval_dim_768_cosine_precision@1": 0.3602150537634409,
|
452 |
+
"eval_dim_768_cosine_precision@10": 0.07300799558864075,
|
453 |
+
"eval_dim_768_cosine_precision@3": 0.1611524676040805,
|
454 |
+
"eval_dim_768_cosine_precision@5": 0.11243451888613179,
|
455 |
+
"eval_dim_768_cosine_recall@1": 0.3602150537634409,
|
456 |
+
"eval_dim_768_cosine_recall@10": 0.7300799558864075,
|
457 |
+
"eval_dim_768_cosine_recall@3": 0.48345740281224153,
|
458 |
+
"eval_dim_768_cosine_recall@5": 0.5621725944306589,
|
459 |
+
"eval_loss": 0.11598330736160278,
|
460 |
+
"eval_runtime": 4756.7643,
|
461 |
+
"eval_samples_per_second": 1.525,
|
462 |
+
"eval_sequential_score": 0.49610418689723323,
|
463 |
+
"eval_steps_per_second": 0.015,
|
464 |
+
"step": 235
|
465 |
},
|
466 |
{
|
467 |
+
"epoch": 6.0,
|
468 |
+
"grad_norm": 89115.0625,
|
469 |
+
"learning_rate": 1.5612549921877228e-06,
|
470 |
+
"loss": 0.0338,
|
471 |
"step": 282
|
472 |
},
|
473 |
{
|
474 |
+
"epoch": 6.0,
|
475 |
+
"eval_dim_128_cosine_accuracy@1": 0.3504273504273504,
|
476 |
+
"eval_dim_128_cosine_accuracy@10": 0.7205679625034463,
|
477 |
+
"eval_dim_128_cosine_accuracy@3": 0.46980976013234077,
|
478 |
+
"eval_dim_128_cosine_accuracy@5": 0.5532120209539564,
|
479 |
+
"eval_dim_128_cosine_map@100": 0.4530896085536035,
|
480 |
+
"eval_dim_128_cosine_mrr@10": 0.44275862031238505,
|
481 |
+
"eval_dim_128_cosine_ndcg@10": 0.5071532407402498,
|
482 |
+
"eval_dim_128_cosine_precision@1": 0.3504273504273504,
|
483 |
+
"eval_dim_128_cosine_precision@10": 0.07205679625034464,
|
484 |
+
"eval_dim_128_cosine_precision@3": 0.15660325337744693,
|
485 |
+
"eval_dim_128_cosine_precision@5": 0.11064240419079129,
|
486 |
+
"eval_dim_128_cosine_recall@1": 0.3504273504273504,
|
487 |
+
"eval_dim_128_cosine_recall@10": 0.7205679625034463,
|
488 |
+
"eval_dim_128_cosine_recall@3": 0.46980976013234077,
|
489 |
+
"eval_dim_128_cosine_recall@5": 0.5532120209539564,
|
490 |
+
"eval_dim_256_cosine_accuracy@1": 0.35594154948993656,
|
491 |
+
"eval_dim_256_cosine_accuracy@10": 0.7259443065894678,
|
492 |
+
"eval_dim_256_cosine_accuracy@3": 0.47615108905431486,
|
493 |
+
"eval_dim_256_cosine_accuracy@5": 0.5562448304383788,
|
494 |
+
"eval_dim_256_cosine_map@100": 0.4584206537452446,
|
495 |
+
"eval_dim_256_cosine_mrr@10": 0.4481468899479667,
|
496 |
+
"eval_dim_256_cosine_ndcg@10": 0.5125341672060905,
|
497 |
+
"eval_dim_256_cosine_precision@1": 0.35594154948993656,
|
498 |
+
"eval_dim_256_cosine_precision@10": 0.07259443065894679,
|
499 |
+
"eval_dim_256_cosine_precision@3": 0.1587170296847716,
|
500 |
+
"eval_dim_256_cosine_precision@5": 0.11124896608767576,
|
501 |
+
"eval_dim_256_cosine_recall@1": 0.35594154948993656,
|
502 |
+
"eval_dim_256_cosine_recall@10": 0.7259443065894678,
|
503 |
+
"eval_dim_256_cosine_recall@3": 0.47615108905431486,
|
504 |
+
"eval_dim_256_cosine_recall@5": 0.5562448304383788,
|
505 |
+
"eval_dim_512_cosine_accuracy@1": 0.35580369451337196,
|
506 |
+
"eval_dim_512_cosine_accuracy@10": 0.7296663909567135,
|
507 |
+
"eval_dim_512_cosine_accuracy@3": 0.4760132340777502,
|
508 |
+
"eval_dim_512_cosine_accuracy@5": 0.5569341053212021,
|
509 |
+
"eval_dim_512_cosine_map@100": 0.4582262561021762,
|
510 |
+
"eval_dim_512_cosine_mrr@10": 0.4481875352843098,
|
511 |
+
"eval_dim_512_cosine_ndcg@10": 0.5133539771570155,
|
512 |
+
"eval_dim_512_cosine_precision@1": 0.35580369451337196,
|
513 |
+
"eval_dim_512_cosine_precision@10": 0.07296663909567135,
|
514 |
+
"eval_dim_512_cosine_precision@3": 0.15867107802591673,
|
515 |
+
"eval_dim_512_cosine_precision@5": 0.11138682106424042,
|
516 |
+
"eval_dim_512_cosine_recall@1": 0.35580369451337196,
|
517 |
+
"eval_dim_512_cosine_recall@10": 0.7296663909567135,
|
518 |
+
"eval_dim_512_cosine_recall@3": 0.4760132340777502,
|
519 |
+
"eval_dim_512_cosine_recall@5": 0.5569341053212021,
|
520 |
+
"eval_dim_64_cosine_accuracy@1": 0.3416046319272126,
|
521 |
+
"eval_dim_64_cosine_accuracy@10": 0.7117452440033085,
|
522 |
+
"eval_dim_64_cosine_accuracy@3": 0.4665012406947891,
|
523 |
+
"eval_dim_64_cosine_accuracy@5": 0.5459057071960298,
|
524 |
+
"eval_dim_64_cosine_map@100": 0.44526802763089235,
|
525 |
+
"eval_dim_64_cosine_mrr@10": 0.43469968402764336,
|
526 |
+
"eval_dim_64_cosine_ndcg@10": 0.4989960354397479,
|
527 |
+
"eval_dim_64_cosine_precision@1": 0.3416046319272126,
|
528 |
+
"eval_dim_64_cosine_precision@10": 0.07117452440033084,
|
529 |
+
"eval_dim_64_cosine_precision@3": 0.1555004135649297,
|
530 |
+
"eval_dim_64_cosine_precision@5": 0.10918114143920596,
|
531 |
+
"eval_dim_64_cosine_recall@1": 0.3416046319272126,
|
532 |
+
"eval_dim_64_cosine_recall@10": 0.7117452440033085,
|
533 |
+
"eval_dim_64_cosine_recall@3": 0.4665012406947891,
|
534 |
+
"eval_dim_64_cosine_recall@5": 0.5459057071960298,
|
535 |
+
"eval_dim_768_cosine_accuracy@1": 0.3615936035290874,
|
536 |
+
"eval_dim_768_cosine_accuracy@10": 0.7302178108629721,
|
537 |
+
"eval_dim_768_cosine_accuracy@3": 0.4807003032809484,
|
538 |
+
"eval_dim_768_cosine_accuracy@5": 0.5638268541494348,
|
539 |
+
"eval_dim_768_cosine_map@100": 0.46347566763660536,
|
540 |
+
"eval_dim_768_cosine_mrr@10": 0.4534180375578229,
|
541 |
+
"eval_dim_768_cosine_ndcg@10": 0.5175916433592258,
|
542 |
+
"eval_dim_768_cosine_precision@1": 0.3615936035290874,
|
543 |
+
"eval_dim_768_cosine_precision@10": 0.07302178108629723,
|
544 |
+
"eval_dim_768_cosine_precision@3": 0.1602334344269828,
|
545 |
+
"eval_dim_768_cosine_precision@5": 0.11276537082988695,
|
546 |
+
"eval_dim_768_cosine_recall@1": 0.3615936035290874,
|
547 |
+
"eval_dim_768_cosine_recall@10": 0.7302178108629721,
|
548 |
+
"eval_dim_768_cosine_recall@3": 0.4807003032809484,
|
549 |
+
"eval_dim_768_cosine_recall@5": 0.5638268541494348,
|
550 |
+
"eval_loss": 0.1151694729924202,
|
551 |
+
"eval_runtime": 4751.4893,
|
552 |
+
"eval_samples_per_second": 1.527,
|
553 |
+
"eval_sequential_score": 0.4989960354397479,
|
554 |
+
"eval_steps_per_second": 0.015,
|
555 |
"step": 282
|
556 |
}
|
557 |
],
|
558 |
"logging_steps": 500,
|
559 |
+
"max_steps": 322,
|
560 |
"num_input_tokens_seen": 0,
|
561 |
+
"num_train_epochs": 7,
|
562 |
"save_steps": 500,
|
563 |
"stateful_callbacks": {
|
564 |
"EarlyStoppingCallback": {
|
|
|
582 |
}
|
583 |
},
|
584 |
"total_flos": 0.0,
|
585 |
+
"train_batch_size": 104,
|
586 |
"trial_name": null,
|
587 |
"trial_params": null
|
588 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5624
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:abcd18cdc1b00a64204792da926e72cd3d5a1ff63da51512062f430b6fa5ad0f
|
3 |
size 5624
|