Update config.json
Browse files- config.json +6 -16
config.json
CHANGED
@@ -24,7 +24,7 @@
|
|
24 |
"task_specific_params": {
|
25 |
"text-generation": {
|
26 |
"do_sample": true,
|
27 |
-
"max_length":
|
28 |
}
|
29 |
},
|
30 |
"transformers_version": "4.34.0",
|
@@ -52,27 +52,17 @@
|
|
52 |
"initial_capacity": 250000,
|
53 |
"input_size": 2048
|
54 |
},
|
55 |
-
"input_dimension":
|
56 |
"low_stability_threshold": 0.01,
|
57 |
-
"max_complexity":
|
58 |
"max_neurons": 250000,
|
59 |
-
"max_sequence_length":
|
60 |
"min_epochs_before_growth": 5,
|
61 |
"model_filename": "pytorch_model.bin",
|
62 |
-
"num_embeddings":
|
63 |
"pruning_improvement_threshold": 0.005,
|
64 |
"some_adaptation_rate": 0.05,
|
65 |
"stability_threshold": 0.02,
|
66 |
"start_token_index": 2
|
67 |
-
}
|
68 |
-
"max_input_length": 2048,
|
69 |
-
"max_total_tokens": 2051,
|
70 |
-
"max_concurrent_requests": 128,
|
71 |
-
"max_best_of": 2,
|
72 |
-
"max_stop_sequences": 4,
|
73 |
-
"max_top_n_tokens": 5,
|
74 |
-
"waiting_served_ratio": 1.2,
|
75 |
-
"max_batch_prefill_tokens": 2048,
|
76 |
-
"max_waiting_tokens": 20,
|
77 |
-
"sharded": false
|
78 |
}
|
|
|
24 |
"task_specific_params": {
|
25 |
"text-generation": {
|
26 |
"do_sample": true,
|
27 |
+
"max_length": 1024
|
28 |
}
|
29 |
},
|
30 |
"transformers_version": "4.34.0",
|
|
|
52 |
"initial_capacity": 250000,
|
53 |
"input_size": 2048
|
54 |
},
|
55 |
+
"input_dimension": 768,
|
56 |
"low_stability_threshold": 0.01,
|
57 |
+
"max_complexity": 10000,
|
58 |
"max_neurons": 250000,
|
59 |
+
"max_sequence_length": 1024,
|
60 |
"min_epochs_before_growth": 5,
|
61 |
"model_filename": "pytorch_model.bin",
|
62 |
+
"num_embeddings": 25000,
|
63 |
"pruning_improvement_threshold": 0.005,
|
64 |
"some_adaptation_rate": 0.05,
|
65 |
"stability_threshold": 0.02,
|
66 |
"start_token_index": 2
|
67 |
+
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
}
|