Commit
•
b2dad91
1
Parent(s):
02620f4
Upload LlamaForCausalLM
Browse files- config.json +15 -15
config.json
CHANGED
|
@@ -6,7 +6,20 @@
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"bos_token_id": 1,
|
| 9 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
"config_groups": {
|
| 11 |
"group_0": {
|
| 12 |
"input_activations": null,
|
|
@@ -36,21 +49,8 @@
|
|
| 36 |
"kv_cache_scheme": null,
|
| 37 |
"quant_method": "compressed-tensors",
|
| 38 |
"quantization_status": "compressed",
|
| 39 |
-
"version": "0.6.0.
|
| 40 |
},
|
| 41 |
-
"eos_token_id": 2,
|
| 42 |
-
"head_dim": 64,
|
| 43 |
-
"hidden_act": "silu",
|
| 44 |
-
"hidden_size": 2048,
|
| 45 |
-
"initializer_range": 0.02,
|
| 46 |
-
"intermediate_size": 5632,
|
| 47 |
-
"max_position_embeddings": 2048,
|
| 48 |
-
"mlp_bias": false,
|
| 49 |
-
"model_type": "llama",
|
| 50 |
-
"num_attention_heads": 32,
|
| 51 |
-
"num_hidden_layers": 22,
|
| 52 |
-
"num_key_value_heads": 4,
|
| 53 |
-
"pretraining_tp": 1,
|
| 54 |
"rms_norm_eps": 1e-05,
|
| 55 |
"rope_scaling": null,
|
| 56 |
"rope_theta": 10000.0,
|
|
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 5632,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 22,
|
| 20 |
+
"num_key_value_heads": 4,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
+
"quantization_config": {
|
| 23 |
"config_groups": {
|
| 24 |
"group_0": {
|
| 25 |
"input_activations": null,
|
|
|
|
| 49 |
"kv_cache_scheme": null,
|
| 50 |
"quant_method": "compressed-tensors",
|
| 51 |
"quantization_status": "compressed",
|
| 52 |
+
"version": "0.6.0.20240928"
|
| 53 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
"rms_norm_eps": 1e-05,
|
| 55 |
"rope_scaling": null,
|
| 56 |
"rope_theta": 10000.0,
|