compact dynatemp into a single value dynatemp_range. This is a float which represents the allowed deviation from the min and max temperature when using dynatemp. Thus, if we want a value of dynatemp_min=0.3, dynatemp_max=0.5, then we would simply set temperature=0.4 and dynatemp_range=0.1. Functionally dynatemp would operate the same, but it would simplify usage and make it a single easy to adjust value.

Author: LostRuins

common/common.h

@@ -82,9 +82,7 @@ struct gpt_params {
     float   mirostat_eta      = 0.10f; // learning rate
 
     // DynaTemp!
-    bool    dynatemp           = false; // enable DynaTemp
-    float   min_temp           = 0.00f; // minimum temperature
-    float   max_temp           = 2.00f; // maximum temperature
+    float   dynatemp_range     = 0.0f;  // enables DynaTemp if greater than 0. dynatemp_min = temperature - dt_range, dynatemp_max = temperature + dt_range
 
     // // sampling parameters
     struct llama_sampling_params sparams;

common/sampling.h

@@ -25,9 +25,7 @@ typedef struct llama_sampling_params {
     int32_t     mirostat              = 0;        // 0 = disabled, 1 = mirostat, 2 = mirostat 2.0
     float       mirostat_tau          = 5.00f;    // target entropy
     float       mirostat_eta          = 0.10f;    // learning rate
-    bool        dynatemp              = false;    // dynamic temperature
-    float       min_temp              = 0.00f; // minimum temperature
-    float       max_temp              = 2.00f; // maximum temperature
+    bool        dynatemp_range        = 0.00f;    // dynamic temperature range
     bool        penalize_nl           = true;     // consider newlines as a repeatable token
     std::string samplers_sequence     = "kfypmt"; // top_k, tail_free, typical_p, top_p, min_p, temp
 

expose.h

@@ -81,9 +81,7 @@ struct generation_inputs
     const char * grammar;
     const bool grammar_retain_state;
     const bool quiet = false;
-    const bool dynatemp = false;
-    const float min_temp;
-    const float max_temp;
+    const float dynatemp_range = 0.0f;
     const logit_bias logit_biases[logit_bias_max];
 
 };

gpttype_adapter.cpp

@@ -481,7 +481,7 @@ void sample_grammar(FileFormat file_format, int32_t n_vocab, llama_token_data_ar
 }
 
 int SampleLogits(const float * logits, int n_ctx, int n_vocab, int rep_pen_range, float rep_pen, float presence_penalty, float top_k, float top_a, float top_p, float min_p, float typical_p, float tfs, float temp, std::mt19937 & rng,
-int mirostat, float mirostat_tau, float mirostat_eta, const std::vector<samplers> & sampler_order, llama_grammar * grammar, bool dynatemp, float min_temp, float max_temp)
+int mirostat, float mirostat_tau, float mirostat_eta, const std::vector<samplers> & sampler_order, llama_grammar * grammar, float dynatemp_range)
 {
     int id = 0;
     std::vector<llama_token_data> candidates;
@@ -540,9 +540,14 @@ int mirostat, float mirostat_tau, float mirostat_eta, const std::vector<samplers
                     llama_sample_typical(nullptr, &candidates_p, typical_p,1);
                     break;
                 case KCPP_SAMPLER_TEMP:
-                    if (dynatemp)
+                    if (dynatemp_range>0)
                     {
-                        llama_sample_entropy(nullptr, &candidates_p, temp, min_temp, max_temp);
+                        float dynatemp_min = temp - dynatemp_range;
+                        float dynatemp_max = temp + dynatemp_range;
+                        //do not allow negative values
+                        dynatemp_min = dynatemp_min<0?0:dynatemp_min;
+                        dynatemp_max = dynatemp_max<0?0:dynatemp_max;
+                        llama_sample_entropy(nullptr, &candidates_p, temp, dynatemp_min, dynatemp_max);
                     }
                     else
                     {
@@ -1502,9 +1507,7 @@ generation_outputs gpttype_generate(const generation_inputs inputs, generation_o
     kcpp_params->mirostat = inputs.mirostat;
     kcpp_params->mirostat_eta = inputs.mirostat_eta;
     kcpp_params->mirostat_tau = inputs.mirostat_tau;
-    kcpp_params->dynatemp = inputs.dynatemp;
-    kcpp_params->min_temp = inputs.min_temp;
-    kcpp_params->max_temp = inputs.max_temp;
+    kcpp_params->dynatemp_range = inputs.dynatemp_range;
     kcpp_params->n_ctx = inputs.max_context_length;
     kcpp_params->n_batch = n_batch;
     kcpp_params->n_threads = n_threads;
@@ -1900,9 +1903,7 @@ generation_outputs gpttype_generate(const generation_inputs inputs, generation_o
             const float presence_penalty = kcpp_params->presence_penalty;
             const float typical_p = kcpp_params->typical_p;
             const float tfs_z = kcpp_params->tfs_z;
-            const float dynatemp = kcpp_params->dynatemp;
-            const float min_temp = kcpp_params->min_temp;
-            const float max_temp = kcpp_params->max_temp;
+            const float dynatemp_range = kcpp_params->dynatemp_range;
 
             if (!startedsampling)
             {
@@ -1958,7 +1959,7 @@ generation_outputs gpttype_generate(const generation_inputs inputs, generation_o
 
             id = SampleLogits(logitsPtr, nctx, n_vocab, last_n_size, repeat_penalty, presence_penalty,
             top_k, top_a, top_p, min_p, typical_p, tfs_z, temp, rng,
-            kcpp_params->mirostat, kcpp_params->mirostat_tau, kcpp_params->mirostat_eta, sampler_order, grammar, dynatemp, min_temp, max_temp);
+            kcpp_params->mirostat, kcpp_params->mirostat_tau, kcpp_params->mirostat_eta, sampler_order, grammar, dynatemp_range);
 
             if (grammar != nullptr) {
                 grammar_accept_token(file_format, n_vocab, grammar, id);

kcpp_docs.embd

@@ -139,18 +139,9 @@
                            "description": "If true, prevents the EOS token from being generated (Ban EOS). For unbantokens, set this to false.",
                            "type": "boolean"
                         },
-                        "dynatemp": {
-                           "default": false,
-                           "description": "If true, uses dynamic temperature. If false, uses static temperature.",
-                           "type": "boolean"
-                        },
-                        "min_temp": {
-                           "description": "Dynatemp Minimum temperature value.",
-                           "exclusiveMinimum": 0,
-                           "type": "number"
-                        },
-                        "max_temp": {
-                           "description": "Maximum temperature value.",
+                        "dynatemp_range": {
+                           "default": 0,
+                           "description": "If greater than 0, uses dynamic temperature. Dynamic temperature range will be between Temp+Range and Temp-Range. If less or equal to 0 , uses static temperature.",
                            "exclusiveMinimum": 0,
                            "type": "number"
                         },