sharpenb's picture
a34344c18ad8ace69523032d44ac72d33d6498dba220e9f1bbb55b2032c42fd9
5f3177c verified
raw
history blame
10 kB
{
"_name_or_path": "/tmp/models/tmp7knn_xpq/tmpmx72f0zj",
"activation_function": "gelu_new",
"all_reduce_scores": {
"0": "NON_PARTICIPATING",
"1": "NON_PARTICIPATING",
"10": "SUCCESS",
"100": "NON_PARTICIPATING",
"101": "NON_PARTICIPATING",
"102": "NON_PARTICIPATING",
"103": "NON_PARTICIPATING",
"104": "NON_PARTICIPATING",
"105": "NON_PARTICIPATING",
"106": "NON_PARTICIPATING",
"107": "SUCCESS",
"108": "NON_PARTICIPATING",
"109": "NON_PARTICIPATING",
"11": "SUCCESS",
"110": "NON_PARTICIPATING",
"111": "SUCCESS",
"112": "NON_PARTICIPATING",
"113": "NON_PARTICIPATING",
"114": "SUCCESS",
"115": "NON_PARTICIPATING",
"116": "NON_PARTICIPATING",
"117": "NON_PARTICIPATING",
"118": "NON_PARTICIPATING",
"119": "SUCCESS",
"12": "SUCCESS",
"120": "SUCCESS",
"121": "NON_PARTICIPATING",
"122": "NON_PARTICIPATING",
"123": "NON_PARTICIPATING",
"124": "NON_PARTICIPATING",
"125": "SUCCESS",
"126": "NON_PARTICIPATING",
"127": "NON_PARTICIPATING",
"128": "SUCCESS",
"129": "SUCCESS",
"13": "SUCCESS",
"130": "NON_PARTICIPATING",
"131": "NON_PARTICIPATING",
"132": "NON_PARTICIPATING",
"133": "NON_PARTICIPATING",
"134": "NON_PARTICIPATING",
"135": "SUCCESS",
"136": "SUCCESS",
"137": "NON_PARTICIPATING",
"138": "NON_PARTICIPATING",
"139": "NON_PARTICIPATING",
"14": "NON_PARTICIPATING",
"140": "NON_PARTICIPATING",
"141": "SUCCESS",
"142": "SUCCESS",
"143": "NON_PARTICIPATING",
"144": "NON_PARTICIPATING",
"145": "SUCCESS",
"146": "NON_PARTICIPATING",
"147": "SUCCESS",
"148": "NON_PARTICIPATING",
"149": "SUCCESS",
"15": "SUCCESS",
"150": "SUCCESS",
"151": "NON_PARTICIPATING",
"152": "SUCCESS",
"153": "NON_PARTICIPATING",
"154": "SUCCESS",
"155": "SUCCESS",
"156": "SUCCESS",
"157": "NON_PARTICIPATING",
"158": "NON_PARTICIPATING",
"159": "NON_PARTICIPATING",
"16": "NON_PARTICIPATING",
"160": "SUCCESS",
"161": "SUCCESS",
"162": "SUCCESS",
"163": "NON_PARTICIPATING",
"164": "SUCCESS",
"165": "NON_PARTICIPATING",
"166": "NON_PARTICIPATING",
"167": "SUCCESS",
"168": "SUCCESS",
"169": "NON_PARTICIPATING",
"17": "NON_PARTICIPATING",
"170": "NON_PARTICIPATING",
"171": "NON_PARTICIPATING",
"172": "SUCCESS",
"173": "SUCCESS",
"174": "SUCCESS",
"175": "NON_PARTICIPATING",
"176": "SUCCESS",
"177": "NON_PARTICIPATING",
"178": "NON_PARTICIPATING",
"179": "NON_PARTICIPATING",
"18": "SUCCESS",
"180": "SUCCESS",
"181": "SUCCESS",
"182": "SUCCESS",
"183": "NON_PARTICIPATING",
"184": "SUCCESS",
"185": "NON_PARTICIPATING",
"186": "NON_PARTICIPATING",
"187": "SUCCESS",
"188": "NON_PARTICIPATING",
"189": "SUCCESS",
"19": "SUCCESS",
"190": "NON_PARTICIPATING",
"191": "SUCCESS",
"192": "SUCCESS",
"193": "NON_PARTICIPATING",
"194": "SUCCESS",
"195": "SUCCESS",
"196": "SUCCESS",
"197": "NON_PARTICIPATING",
"198": "NON_PARTICIPATING",
"199": "NON_PARTICIPATING",
"2": "NON_PARTICIPATING",
"20": "SUCCESS",
"200": "NON_PARTICIPATING",
"201": "NON_PARTICIPATING",
"202": "NON_PARTICIPATING",
"203": "SUCCESS",
"204": "NON_PARTICIPATING",
"205": "NON_PARTICIPATING",
"206": "NON_PARTICIPATING",
"207": "NON_PARTICIPATING",
"208": "SUCCESS",
"209": "NON_PARTICIPATING",
"21": "SUCCESS",
"210": "NON_PARTICIPATING",
"211": "NON_PARTICIPATING",
"212": "NON_PARTICIPATING",
"213": "NON_PARTICIPATING",
"214": "SUCCESS",
"215": "SUCCESS",
"216": "SUCCESS",
"217": "NON_PARTICIPATING",
"218": "NON_PARTICIPATING",
"219": "SUCCESS",
"22": "NON_PARTICIPATING",
"220": "NON_PARTICIPATING",
"221": "NON_PARTICIPATING",
"222": "NON_PARTICIPATING",
"223": "NON_PARTICIPATING",
"224": "NON_PARTICIPATING",
"225": "NON_PARTICIPATING",
"226": "NON_PARTICIPATING",
"227": "SUCCESS",
"228": "SUCCESS",
"229": "NON_PARTICIPATING",
"23": "SUCCESS",
"230": "SUCCESS",
"231": "NON_PARTICIPATING",
"232": "SUCCESS",
"233": "NON_PARTICIPATING",
"234": "NON_PARTICIPATING",
"235": "NON_PARTICIPATING",
"236": "NON_PARTICIPATING",
"237": "NON_PARTICIPATING",
"238": "NON_PARTICIPATING",
"239": "NON_PARTICIPATING",
"24": "SUCCESS",
"240": "SUCCESS",
"241": "NON_PARTICIPATING",
"242": "SUCCESS",
"243": "SUCCESS",
"244": "NON_PARTICIPATING",
"245": "SUCCESS",
"246": "NON_PARTICIPATING",
"247": "NON_PARTICIPATING",
"248": "SUCCESS",
"249": "NON_PARTICIPATING",
"25": "NON_PARTICIPATING",
"250": "NON_PARTICIPATING",
"251": "SUCCESS",
"252": "SUCCESS",
"253": "NON_PARTICIPATING",
"254": "SUCCESS",
"255": "NON_PARTICIPATING",
"26": "NON_PARTICIPATING",
"27": "NON_PARTICIPATING",
"28": "SUCCESS",
"29": "NON_PARTICIPATING",
"3": "NON_PARTICIPATING",
"30": "NON_PARTICIPATING",
"31": "NON_PARTICIPATING",
"32": "NON_PARTICIPATING",
"33": "NON_PARTICIPATING",
"34": "NON_PARTICIPATING",
"35": "SUCCESS",
"36": "NON_PARTICIPATING",
"37": "SUCCESS",
"38": "NON_PARTICIPATING",
"39": "NON_PARTICIPATING",
"4": "NON_PARTICIPATING",
"40": "NON_PARTICIPATING",
"41": "NON_PARTICIPATING",
"42": "SUCCESS",
"43": "SUCCESS",
"44": "NON_PARTICIPATING",
"45": "SUCCESS",
"46": "NON_PARTICIPATING",
"47": "SUCCESS",
"48": "NON_PARTICIPATING",
"49": "SUCCESS",
"5": "NON_PARTICIPATING",
"50": "SUCCESS",
"51": "NON_PARTICIPATING",
"52": "NON_PARTICIPATING",
"53": "SUCCESS",
"54": "NON_PARTICIPATING",
"55": "NON_PARTICIPATING",
"56": "NON_PARTICIPATING",
"57": "SUCCESS",
"58": "NON_PARTICIPATING",
"59": "NON_PARTICIPATING",
"6": "SUCCESS",
"60": "NON_PARTICIPATING",
"61": "SUCCESS",
"62": "NON_PARTICIPATING",
"63": "NON_PARTICIPATING",
"64": "SUCCESS",
"65": "NON_PARTICIPATING",
"66": "NON_PARTICIPATING",
"67": "NON_PARTICIPATING",
"68": "NON_PARTICIPATING",
"69": "NON_PARTICIPATING",
"7": "SUCCESS",
"70": "NON_PARTICIPATING",
"71": "SUCCESS",
"72": "SUCCESS",
"73": "SUCCESS",
"74": "SUCCESS",
"75": "SUCCESS",
"76": "NON_PARTICIPATING",
"77": "NON_PARTICIPATING",
"78": "NON_PARTICIPATING",
"79": "NON_PARTICIPATING",
"8": "SUCCESS",
"80": "NON_PARTICIPATING",
"81": "SUCCESS",
"82": "NON_PARTICIPATING",
"83": "SUCCESS",
"84": "SUCCESS",
"85": "NON_PARTICIPATING",
"86": "NON_PARTICIPATING",
"87": "NON_PARTICIPATING",
"88": "SUCCESS",
"89": "NON_PARTICIPATING",
"9": "NON_PARTICIPATING",
"90": "NON_PARTICIPATING",
"91": "NON_PARTICIPATING",
"92": "SUCCESS",
"93": "SUCCESS",
"94": "NON_PARTICIPATING",
"95": "NON_PARTICIPATING",
"96": "SUCCESS",
"97": "NON_PARTICIPATING",
"98": "NON_PARTICIPATING",
"99": "NON_PARTICIPATING"
},
"architectures": [
"GPTOptim"
],
"attn_pdrop": 0.1,
"auto_map": {
"AutoConfig": "configuration_gpt_optimized.GPTOptimConfig",
"AutoModelForCausalLM": "distributed/optimized-gpt2-2b--modeling_gpt_optimized.GPTOptim"
},
"block_size": 1024,
"bos_token_id": 50256,
"embd_pdrop": 0.1,
"eos_token_id": 50256,
"initializer_range": 0.02,
"layer_norm_epsilon": 1e-05,
"model_type": "gpt_optimized",
"n_embd": 1904,
"n_head": 34,
"n_inner": null,
"n_layer": 44,
"n_positions": 1024,
"quantization_config": {
"transformer.h.attn.c_attn": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.attn.c_proj": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.mlp.c_fc": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
},
"transformer.h.mlp.c_proj": {
"offload_meta": false,
"scale_quant_params": null,
"weight_quant_params": {
"axis": 0,
"channel_wise": true,
"group_size": 64,
"nbits": 8,
"optimize": true,
"round_zero": false,
"view_as_float": false
},
"zero_quant_params": {
"channel_wise": false,
"group_size": null,
"nbits": 8,
"optimize": false
}
}
},
"reorder_and_upcast_attn": false,
"resid_pdrop": 0.1,
"scale_attn_by_inverse_layer_idx": false,
"scale_attn_weights": true,
"summary_activation": null,
"summary_first_dropout": 0.1,
"summary_proj_to_labels": true,
"summary_type": "cls_index",
"summary_use_proj": true,
"torch_dtype": "float32",
"transformers_version": "4.44.0",
"use_cache": true,
"vocab_size": 50257
}