Upload LlamaForCausalLM
Browse files- config.json +3 -2
- generation_config.json +1 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "/gpfs/projects/bsc88/text/models/salamandra_quantized_speculative/salamandra-7b-instruct-fp8",
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
@@ -47,6 +47,7 @@
|
|
| 47 |
"quantization_status": "frozen"
|
| 48 |
},
|
| 49 |
"eos_token_id": 2,
|
|
|
|
| 50 |
"hidden_act": "silu",
|
| 51 |
"hidden_size": 4096,
|
| 52 |
"initializer_range": 0.02,
|
|
@@ -63,7 +64,7 @@
|
|
| 63 |
"rope_theta": 10000.0,
|
| 64 |
"tie_word_embeddings": false,
|
| 65 |
"torch_dtype": "float32",
|
| 66 |
-
"transformers_version": "4.
|
| 67 |
"use_cache": true,
|
| 68 |
"vocab_size": 256000
|
| 69 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/home/jsaiz/storage5/gpfs/projects/bsc88/text/models/salamandra_quantized_speculative/salamandra-7b-instruct-fp8",
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
|
|
| 47 |
"quantization_status": "frozen"
|
| 48 |
},
|
| 49 |
"eos_token_id": 2,
|
| 50 |
+
"head_dim": 128,
|
| 51 |
"hidden_act": "silu",
|
| 52 |
"hidden_size": 4096,
|
| 53 |
"initializer_range": 0.02,
|
|
|
|
| 64 |
"rope_theta": 10000.0,
|
| 65 |
"tie_word_embeddings": false,
|
| 66 |
"torch_dtype": "float32",
|
| 67 |
+
"transformers_version": "4.46.1",
|
| 68 |
"use_cache": true,
|
| 69 |
"vocab_size": 256000
|
| 70 |
}
|
generation_config.json
CHANGED
|
@@ -8,5 +8,5 @@
|
|
| 8 |
],
|
| 9 |
"repetition_penalty": 1.2,
|
| 10 |
"temperature": 0.6,
|
| 11 |
-
"transformers_version": "4.
|
| 12 |
}
|
|
|
|
| 8 |
],
|
| 9 |
"repetition_penalty": 1.2,
|
| 10 |
"temperature": 0.6,
|
| 11 |
+
"transformers_version": "4.46.1"
|
| 12 |
}
|