Upload LlamaForCausalLM
Browse files- config.json +73 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/gpfs/projects/bsc88/text/models/salamandra_quantized_speculative/V1.1/salamandra-2b-instruct-fp8",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 5440,
|
| 15 |
+
"max_position_embeddings": 8192,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"num_key_value_heads": 16,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
+
"quantization_config": {
|
| 23 |
+
"config_groups": {
|
| 24 |
+
"group_0": {
|
| 25 |
+
"input_activations": {
|
| 26 |
+
"actorder": null,
|
| 27 |
+
"block_structure": null,
|
| 28 |
+
"dynamic": true,
|
| 29 |
+
"group_size": null,
|
| 30 |
+
"num_bits": 8,
|
| 31 |
+
"observer": null,
|
| 32 |
+
"observer_kwargs": {},
|
| 33 |
+
"strategy": "token",
|
| 34 |
+
"symmetric": true,
|
| 35 |
+
"type": "float"
|
| 36 |
+
},
|
| 37 |
+
"output_activations": null,
|
| 38 |
+
"targets": [
|
| 39 |
+
"Linear"
|
| 40 |
+
],
|
| 41 |
+
"weights": {
|
| 42 |
+
"actorder": null,
|
| 43 |
+
"block_structure": null,
|
| 44 |
+
"dynamic": false,
|
| 45 |
+
"group_size": null,
|
| 46 |
+
"num_bits": 8,
|
| 47 |
+
"observer": "minmax",
|
| 48 |
+
"observer_kwargs": {},
|
| 49 |
+
"strategy": "channel",
|
| 50 |
+
"symmetric": true,
|
| 51 |
+
"type": "float"
|
| 52 |
+
}
|
| 53 |
+
}
|
| 54 |
+
},
|
| 55 |
+
"format": "float-quantized",
|
| 56 |
+
"global_compression_ratio": 1.4564469860437392,
|
| 57 |
+
"ignore": [
|
| 58 |
+
"lm_head"
|
| 59 |
+
],
|
| 60 |
+
"kv_cache_scheme": null,
|
| 61 |
+
"quant_method": "compressed-tensors",
|
| 62 |
+
"quantization_status": "compressed",
|
| 63 |
+
"sparsity_config": {}
|
| 64 |
+
},
|
| 65 |
+
"rms_norm_eps": 1e-05,
|
| 66 |
+
"rope_scaling": null,
|
| 67 |
+
"rope_theta": 10000.0,
|
| 68 |
+
"tie_word_embeddings": false,
|
| 69 |
+
"torch_dtype": "float16",
|
| 70 |
+
"transformers_version": "4.49.0",
|
| 71 |
+
"use_cache": true,
|
| 72 |
+
"vocab_size": 256000
|
| 73 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
2,
|
| 7 |
+
5
|
| 8 |
+
],
|
| 9 |
+
"repetition_penalty": 1.2,
|
| 10 |
+
"temperature": 0.6,
|
| 11 |
+
"transformers_version": "4.49.0"
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b23571ea329b1a98f9316c1655d446ce5cd337b8de80a37964775773a0678051
|
| 3 |
+
size 3303225952
|